Presentation
The university hospital is a model of excellence in Italy for pre-clinical, translational, and clinical research and care activities. It is equipped with 110 beds to treat all types of oncological pathologies in adults. There are 115 researchers working there. The hospital is structured into six departments, of which three are clinical (Medical Area, Diagnosis and Imaging Therapy, Surgical Area), two are services, and one is an administrative/management department. The medical area includes four wards: medical oncology for thoracic pathology, medical oncology, heamatology, medical oncology for oncology patient care. In 2015, the institute was accredited as a clinical cancer centre according to the Organization of European Cancer Institutes (OECI). Since 2015, evidence-based medicine and patient-centred care methodologies have been successfully implemented in the hospital, but no process improvement methodology has been used. In 2017, the hospital became a hub for oncological diseases, which led to an increased demand for care and services. The hospital has received national funds dedicated to hubs and has made investments in infrastructure improvements and the purchase of new innovative medical equipment.
a) Contextual factors enabling or hindering lean introduction
The description of the external and internal contextual factors, as revealed in the first questionnaire and the interviews, is given in Table 1. Below is a brief description of each item.
External context and organizational elements driving Lean Introduction in the haematology ward
The analysis of the context revealed external and internal elements influencing the introduction of lean. Starting with the external elements, the most frequently discussed motivators that led to the search for methodologies for process improvement include the continuous increase in patient volume and the benchmark of process performance with other providers. Although the clinical results were above the national average, the increase in demand - especially in the medical area - highlighted the inability to manage the increasing flow of patients. The inability to manage the increasing number of patients also affected the performance of the process in the diagnostic area.
Table 1 External and internal contextual factors recognized by hospital staff.
External Context
|
|
Increased demand for service
|
|
Process performance lower than benchmark
|
Lean introduction drivers
|
|
Patients dissatisfied with waiting time and length of stay
|
|
Bottlenecks in the diagnostic area, UMACA, and haematochemical analysis laboratories
|
|
General Manager and Clinical Director sponsorship
|
Hindering elements
|
|
Distrust from doctors and managers of many departments in the medical area
|
|
Time-resource
|
|
Conflicts between hospital and medical area managers
|
|
Lack of knowledge of lean methodology
|
|
Zero maturity level in lean methodology
|
|
Lack of process improvement culture
|
Enabling elements in haematology ward
|
|
Motivation of haematology manager and physicians
|
|
Commitment of doctors and nurses in the haematology ward
|
|
Decision-making and authority to the chief medical physician
|
|
Training focused on team management and learning in practice
|
|
Training and supervision of an experienced consultant
|
|
Background knowledge of other patient-focused approaches and improvement of clinical outcomes
|
|
Motivations related to the correct definition of the ward layout and its organization
|
Internal elements driving the lean introduction were related to dissatisfaction with inefficient work practices within the medical area and the dissatisfaction of many patients who complained about long wait times and lengths of stay.
The choice of lean methodology derives from the desire to follow the example of certain Tuscan hospitals that have been using lean at a strategic level since 2015. These hospitals are considered the benchmark for continuous process improvement. In addition, the methodology was strongly sponsored by the clinical director and the general director of the hospital. They had participated in a 60-hour regional training course on lean healthcare in the second half of 2017. During the training course, they studied case studies of excellence in lean implementation.
When, in May 2018, the CEO proposed the introduction of lean methodology in the medical area, the head physicians showed strong resistance because of the resources that would need to be allocated to the implementation process. In addition, some doctors did not trust the method. This brought up some conflicts with the medical area managers. The haematology staff, represented by their head physician, were the only ones who explicitly agreed to implement the lean introduction. The department, as in most Italian hospitals, is structured as a clinical area where the physicians —in contrast to other professionals— were members of the ward organizationally. Haematology staff were strongly motivated to do research and achieve excellent process performance. They were interested in taking the opportunity to define excellent clinical pathways, as the ward was undergoing managerial and layout restructuring. In addition, the haematology staff believed that lean could further improve clinical performance and improve the patient-centred and evidence-based approach. Until mid-2017 the ward was part of oncology; afterward, it was made independent and new areas of the hospital were assigned to it. Since the ward became independent, one head physician, three doctors and four nurses have been hired. The department is equipped with the most modern medical equipment. The layout of the ward was not yet fully defined, and some rooms that could have potentially been assigned to medical, diagnostic and therapeutic activities had not been assigned to process activities. The ward shares the Antiblastic Chemotherapy Handling Unit (UMACA) and the analysis laboratory with the other four medical department wards in the hospital, so the staff needed to coordinate clinical processes so as not to create bottlenecks.
Since haematology is a strategic ward for the hospital, and in the last two years the demand for treatment has increased more than in other wards, the managers of the medical area have deemed it appropriate to introduce lean there. Moreover, although the clinical pathways were characterized by excellent clinical outcomes, benchmarking activities showed that LOS and MT performance was lower than the performance of the four hospitals chosen as the basis for comparison (Table 2).
Internal contextual elements enabling and hindering lean introduction in the haematology ward
At the organizational level, hospital management has strongly supported the introduction of the method. Since the haematology staff had no experience in process improvement activities, management provided the budget for an external consultant. In addition, three non-clinical personnel were allocated part-time to support the implementation of visual management systems and communication. The organizational structure of the ward has been modified to a matrix form. A team of three haematology ward physicians and two nurses was established and the ward’s head physician was elected project manager. The project manager had formal authority over the team and the personnel employed in the process to be improved; this reduced conflicts due to the double line of hierarchical authority. The lean advisor supported the group for eight months through training and project supervision. He coordinated two meetings per week and carried out Kata coaching activities. The theoretical training activity, lasting five week ends (in June 2018), was differentiated to accommodate technical and managerial competency needs. The team project manager and the medical area manager were trained on topics such as project management, team management, leadership, and the dissemination of lean. The members of the lean group were trained in lean techniques and tools. The key principles of lean thinking, the PDCA cycle methodology and lean assessment were taught to all participants. The most difficult barrier to overcome was the time available. The team agreed to spend eight hours per week on training and pilot project implementation. The management of the team was facilitated by the experience gained with the implementation of the patient-centred care and evidence-based medicine. The motivation of the medical staff–microsystem element–and the focus on team management were key success factors for the involvement of team members. The culture of change introduced by patient-centred and evidence-based medicine was another enabling factor.
b) Implementation strategies
Pilot project definition
Hospital managers and lean team members, who had experience in implementing patient-centred care methodologies, suggested starting a pilot project for the lean introduction. The consultant agreed. The team, with the support of the expert, analysed the clinical pathways in haematology. Six pathways emerged: a) diagnostic visits, b) biopsies, c) check-up visits, d) transfusions, e) infusion chemotherapies, and f) oral chemotherapies. Hospital managers argued that the pathway of the patient undergoing infusion chemotherapy was the most critical for patient and organization value. This process is the only one that involves several departments and requires a large amount of materials and time-consuming resources. In the first and second half of 2017 and 2018, there was a significant increase in the number of chemotherapeutic preparations. LOS, P3, and MT performance decreased during the same periods (Table 2). In addition, outpatient visits and the number of biopsies also increased. The medical staff stated that the increase in demand in the medical area had particularly affected the infusion therapy activities because they involved technical and instrumental resources that are shared with other departments (Table 2). The length of stay was analyzed for patients undergoing short (LOSs) and long-term infusion (LOSl) chemotherapy. The first has a minimum duration of 90 minutes and a maximum of 180 minutes, and the second has a minimum duration of 181 minutes and a maximum of 360 minutes. Each patient was assigned to one of the infusion treatment classes. Process data were collected and analysed by the Department Management Control Office. The process performance data collection and reports were established in 2015 for the implementation of evidence-based medicine.
Pilot project implementation
The pilot project started in June 2018. The first month was dedicated to Gemba Walk, Methods-Time Measurement (MTM) and implementation of the 5S. In addition, the consultant trained the project manager, department managers and lean team members. There were many difficulties during the training period, especially with regard to process mapping and the concept of value, the latter being interpreted by doctors as clinical output. The non-medical staff dedicated to the project assisted the team in the drawing of the visual management material. A room in the medical department was dedicated for team meetings, and some notice boards were installed to post the materials developed during the project. The project activities were organized according to the Report A3 scheme. It followed the phases of the consolidated Deming cycle: Plan-Do-Check-Act (PDCA). Implementing the approach proposed by Deming allowed for the trial-and-error empirical method to be abandoned in favor of the "scientific" one. The PDCA allowed accurate planning of objectives and activities and their monitoring. The departmental managers and the consultant through the study of the national publications and explicit requests to colleagues in other hospitals - considered virtuous - identified the benchmark (Table 2). They took into account the hospital's specific characteristics, such as the policy of not accepting haematochemical reports from outside for fragile patients. This choice is dictated by the risk management plan and affects P3 and MT performance. Time for blood sampling and haematochemical analysis is added to the cycle time; however, it eliminates many risks associated with clinical treatment.
Table 2 Performance indicators of cancer patient pathways in the haematology ward with semiannual variation and benchmark.
Period
|
I-2017
|
II-21017
|
I-2018
|
II-2018
|
Benchmark
|
Total
|
Total
|
∆
|
Total
|
∆
|
Total
|
∆
|
|
|
Outpatient visits
|
5760
|
6202
|
442
|
6714
|
512
|
6686
|
-28
|
|
|
Infusion Chemotherapy
|
1644
|
1752
|
108
|
1844
|
92
|
1915
|
71
|
|
|
No. of Chemo Chairs
|
7
|
7
|
0
|
8
|
1
|
8
|
0
|
|
|
Performance
|
Average
|
Average
|
∆
|
Average
|
∆
|
Average
|
∆
|
Average
|
∆
|
MT [#]
|
2.28
|
2.09
|
-0.19
|
1.88
|
-0.21
|
2.11
|
0,23
|
2.50
|
0.39
|
P3
|
37.50%
|
36.20%
|
-1.30%
|
35.80%
|
-0.40%
|
36.00%
|
0.20%
|
70%
|
34.0%
|
LOS
|
|
|
|
|
|
|
|
|
|
LOSs [min]
|
302
|
318
|
16
|
344
|
26
|
328
|
-16
|
250
|
-78
|
LOSl [min]
|
414
|
421
|
7
|
447
|
26
|
427
|
-20
|
330
|
-97
|
The existing care process was mapped through Value Stream Mapping (VSM) based on the patient reports, Gemba walks, interviews, and direct observation. For instance, Figure 3 shows the pathway of a patient undergoing short-term infusion chemotherapy. The cycle time in Figure 3 was calculated over an observation period of one week and included 51 patients. In addition, the application of the Demand Map and the Spaghetti Chart were used to evaluate the ward nodes activated by the patients and the ward’s layout. These tools were useful in defining the possible sources of waste in the process. The application of these tools lasted more than two months and required several revisions. Once completed, the results were posted in the meeting room and were used for discussions with colleagues in the medical department. The lean team requested support from the consultant for the drafting of the VSM and for the layout analysis. In addition, the consultant was asked to simplify negotiations with staff from other departments who were reluctant to be subjected to time and method measurements. During the planning phase, many difficulties emerged, including: selection of a standard of measurement, coordination of work and meeting schedules, and the deadlines set by the project Gantt. Although the project manager was able to manage the team, he did not have enough experience in lean tools. The external consultant played a key role in managing these activities.
At the end of the as-is analysis process, an Ishikawa diagram was used for the definition of root causes. Four root causes emerged from the meetings and interviews. They were patient flow management, coordination activities with other departments, layout, and Information Technology equipment (IT).
Patient flow management concerned the absence of priority in the management of patients based on the clinical path and the arrival of patients in the early hours of the morning. The lack of coordination with other departments led to delays in the preparation of infusion chemotherapy and blood test reports. The layout was such that the flow of doctors and nurses crossed the flow of patients, and this caused great inconvenience to the doctors and nurses. Also, the computer softwares was not compatible, which meant that the same data had to be recorded several times.
After some meetings and rigorous brainstorming, the lean team suggested changes to be made in the existing pathway. This was done by considering how patients could be divided into batches so that the infusion activity could start as soon as possible without affecting other activities and by simplifying coordination with the other diagnostic units. Patient lot sizes were calculated so that long-term patients are given priority and short-term patients are treated in a way that limits wait times and does not affect the other wards’ activities. Theories of queues and operational research methodologies were implemented to address chemo chair saturation. A chemo chair activities plan was implemented through pull logic.
In addition, the hospital engineer was involved in making sure the information systems were compatible. Whenever integrating the software was not possible, a data entry person was assigned to prevent medical staff from wasting their time on low-value activities. The ward layout has been modified to prevent patient flows from intersecting with the flows of doctors and nurses. In addition, the use of one room has been changed from a small warehouse to a blood collection room to increase the value of the activities carried out within it. The waiting rooms were moved outside the ward and, during the first two hours of the working day, the biopsy room was reassigned to blood collection activities to speed up the requests for therapies in UMACA. Patient intake, blood collection, and tube labeling activities have been paralleled to be performed simultaneously in the same room. The routes and modalities for the delivery of blood samples to the laboratories were revised in order to reduce the time and distance traveled by non-clinical staff. Tablet reporting systems were installed. Finally, a patient chemo chair allocation system was developed.
The resources needed for these changes were determined. The team tested and modified the changes during December 2018 and January 2019. The tests were evaluated based on the performance data, patient reports and the team’s expertise.
c) Pilot project results
In January 2019, it was decided to implement the new standard procedures that were tested in order to improve performance. The team met once a week for six months. On a monthly basis, performance was reviewed and new changes were tested. Clinical and nonclinical personnel from other wards and departments were invited to each weekly meeting to share with them the results of the pilot project, and to involve them in the lean methodology.
Every morning, the team leader investigated the impact of organizational changes in order to avoid conflicts. Organizational problems that emerged were discussed and resolved by consensus. In the follow-up phase, the consultant performed supervisory activities. Each week, the team leader performed the Kata coaching. During the first six months, the monitoring of activity was very frequent to prevent a return to old operating modes. Subsequently, when the staff had learned the new procedures, monitoring was reduced to once a month.
Table 3 shows the results achieved through the implementation of the pilot project. The benchmark was not reached for all indicators; however, the results improved over time. The MT indicator is lower by 2.5 due to the number of patients undergoing long-term therapy. Figure 4 shows the to-be state of the same process analysed in Figure 3. From the cycle time analysis of each process step, the areas of waste eliminated are clear.
The incremental improvements in process performance over time are explained by the need for staff to learn new procedures in the early period. In addition, the patients’ resistance to changing their habits also slowed down the improvement in performance. Patients have been educated over time, through an intense communication activity based on visual management systems and telephone reminders.
Table 3 Pilot project results in the follow-up period and act phase.
Period
|
II-2018
|
I-2019
|
II-2019
|
I-2020*
|
Benchmark
|
Total
|
Total
|
∆
|
Total
|
∆
|
Total
|
∆
|
|
|
Outpatient visits
|
6686
|
6944
|
258
|
7005
|
65
|
5629
|
-1073
|
|
|
Infusion Chemotherapy
|
1915
|
2037
|
122
|
2049
|
12
|
1716
|
-330
|
|
|
Number of Chemo Chairs
|
8
|
8
|
0
|
8
|
0
|
8
|
0
|
|
|
Performance
|
Average
|
Average
|
∆
|
Average
|
∆
|
Average**
|
∆
|
Average
|
|
MT [#]
|
2.11
|
2.42
|
0.31
|
2.46
|
0.04
|
2.46
|
0.00
|
2.50
|
0.04(b)
|
P3
|
36.00%
|
58.4%
|
22.4%
|
65.9%
|
11.5%
|
68.2%
|
2.3%
|
70%
|
1.8(b)
|
LOS
|
|
|
|
|
|
|
|
|
|
LOSs [min]
|
328
|
274
|
-54
|
261
|
-13
|
254
|
-7
|
250
|
4(b)
|
LOSl [min]
|
427
|
341
|
-86
|
333
|
-8
|
327
|
-6
|
330
|
-3(a)
|
* Data for the first 5 months of 2020
|
(a) Performance level achieved
|
** 5-month performance average
|
(b) Performance level not achieved
|
In addition to the results showed in Table 3, the pilot project had a positive impact on the performance of other patient pathways in the medical department. The cycle time variability reduction and the leveling of the service demand allowed the UMACA and the analysis laboratory to better plan their activities. The new layout reduced waste due to unnecessary movement. Nurses walk 2 km less per day and doctors 1.5 km less per day. Software integration has reduced data logging time by 35 minutes per day for each doctor. Patients have evaluated the change positively. In particular, they have experienced a drastic reduction in wait times, and greater attention from the medical staff. Increased privacy and a precise time of service are other improvements reported by patients.
Finally, the clinical staff was satisfied with the new procedures because they reduce overloads and allowed for better planning of activities. They say that dividing patients into time slots based on clinical priority reduces stress and simplifies the coordination of activities with other departments. In June 2019 the results were celebrated with a formal team award ceremony. The resulting Report A3 was posted on the bulletin boards in the hospital wards and in the reception area.
d) Lean dissemination and adoption in the hospital
In June 2019, members of the pilot project lean team were promoted to the position of lean sponsors. Their role was to disseminate the lean methodology in the medical department. Following the success of the pilot project, the hospital managers set up a Lean Support Office and assigned to it the three non-clinical resources that had already supported the pilot project. The change of context and enabling factors were of great importance at this stage. The drive to implement lean was characterized by both the need to improve process performance and to increase trust in the method. The desire emulate the success of the pilot project prompted the doctors in charge of the other departments in the medical area to seek the support of the consultants to implement process improvement pathways. Hospital management provided peer internal training hours and days. The first methodology to spread throughout the medical department was 5S. According to the lean sponsors, this methodology was a fundamental condition for implementing lean methodologies. In addition, visual management systems have been implemented to facilitate change. In August 2019, three projects were undertaken in the medical oncology for thoracic pathology and the medical oncology wards. Two of them concerned the same clinical pathway addressed in the pilot project, and the last one was the harmonization of protocols for caring for an oncological patient between departments. Each project has been implemented following the A3 report scheme with the support of the lean sponsors, who were assigned the role of project management. Teams of three doctors and one nurse were dedicated to each project. The projects were shorter in duration than the pilot project because of the maturity achieved. Although the peer training and project management being carried out by a doctor simplified the negotiation in some cases, organizational and structural barriers emerged. The difficulty of getting the new procedures accepted, the impossibility of optimizing the layouts and the respect of the project manager's authority limited the improvement in performance.
In September 2019 the diagnostic department started 5S and visual management implementation initiatives. In October 2019 the same initiatives were undertaken in the surgical department. These initiatives were spontaneously implemented. The managers of these departments have asked the hospital director to introduce lean in their departments. Given the maturity of the method and the number of doctors trained, hospital managers did not consider it possible to undertake systemic improvement pathways in all departments. However, they have changed the organizational structures of the departments into matrix structures. Two doctors with lean experience, per department, have been assigned the role of project manager. The project managers have sponsored peer training and Kaizen blitz activities throughout the hospital departments. In the period October to December 2019 more than 60 doctors and nurses were trained in 40-hour courses by their colleagues, and three Kaizen blitz projects in the diagnostic department and two Kaizen blitz projects in the surgical area were carried out.
In December 2020, in all the departments discussed so far, doctors were involved in continuous improvement activities, with projects structured through the use of both PDCA and Kaizen blitz. The activities were undertaken spontaneously without the supervision of a manager and without any impact on daily clinical activity. The maturity of the methodology, the support of colleagues, and trust were enabling elements. However, some barriers such as infrastructural constraints, and coordination of doctors and nurses and information systems have frequently affected the implementation of the method.
Due to the success of implementations at the micro level, managers have attempted to implement the lean methodology at the meso level. Hospital managers discussed, formalized and communicate in organization the Lean Strategic Plan. In January 2020, the Lean Support Office was transformed into a lean projects control room and renamed as the Operation Management Office. The role of this office is to define lean development policies and to supervise continuous improvement activities. The office has been placed in line with the strategic direction. Two lean project managers, two hospital managers, and three administrative officers have been assigned to it. Lean assessment, to evaluate the knowledge of lean in organization, and Honshi Kanri, to strategically govern change activities, were implemented at the organizational level. These two tools did not seem to provide the expected results because the office project managers did not always agree with the hospital management on the priority of the projects. In addition, there are often disagreements between the operations management office and departmental project managers about when to start a project and how to manage it and communicate project results. Although there are many process improvement projects underway, these have not always been decided harmoniously between the Operation Management Office and the hospital departments. Although many improvement actions take place within the departments, the operation management office is unable to govern the continuous improvement processes. Medical leadership seems to prevail over managerial leadership; therefore, there is a strong difficulty in strategically governing continuous improvement.
After the pilot project and the initial push for implementation by management, internal contextual factors changed radically within the organization. While initially sponsorship and management involvement were necessary for lean implementation, today the methodology is independently disseminated. In particular, small improvement groups have emerged that are able to address various challenges. Process vision and patient focus have become part of the hospital culture. Doctors claim that continuous improvements simplify daily work, save time, and increase the level of service and the number of services provided.