2. Materials and Methods
2.1 Theoretical Framework
The framework for this improvement work was a quasi-experimental, single-study site, pretest-posttest study which was conducted from September 2022 to May 2023. Quasi experimental refers to a design type used to evaluate interventions [18]. Quasi-experimental design facilitates the use of both preintervention and postintervention measurements within process improvement studies, to measure the occurrence of an outcome before and after a particular intervention is implemented [18]. We were aware that pre and post-intervention design has the limitation of ascribing with certainty results to an intervention [19], but it has been used widely to evaluate Lean Six Sigma improvement projects in healthcare [20-22]. We therefore felt it suitable to use Lean Six Sigma methodology within a quasi-experimental pre-and post-intervention study design framework. It was explicit in our study design that any outcomes from redesigning the referral process could then be extrapolated and applied to other clinics as necessitated.
2.2 Use of a Person-centred Lean Six Sigma model
As outlined, the project team had decided to underpin their improvement work with person-centred principles (section 1.5). Person-centred approaches have an explicit focus on ensuring the client or patient is at the center of care delivery but are also concerned with every person involved in the patient’s care, including their families and the staff providing the care [23]. More recent studies have demonstrated that Lean Six Sigma and Person-centred care have been shown to be synergistic in their approach to eliciting the customer voice, understanding and delivering on customer needs [24,25]. These studies have led to the development of a combined model of both Person-centred and Lean Six approaches for use in healthcare [26] which acts as a guide for improvement practitioners in delivering Person-centred Lean Six Sigma improvement [23]. In use the model has shown to have contributed positively to solving problems in both inpatient and outpatient settings [27, 28] and more recently in both acute hospital and community ophthalmology services in Ireland [13]. The model pays particular attention to the synergistic elements of both Person-centred approaches and Lean Six Sigma, in particular:
- Voice of the Customer
- Observational Studies
- Respect for Person
- Staff empowerment
Our work was therefore informed and underpinned by the use of a combined Person-centered Lean Six Sigma approach [13,23] using the Person-centred Lean Six Sigma model [26.]
Specific examples of Lean Six Sigma improvements in healthcare include:
- Reduced wait times and faster access to treatment in Emergency Departments [29];
- Improved patient outcomes in Cardiac Units [30];
- Streamlined nursing drug rounds on wards [31];
2.2 Design
Lean Six Sigma methodology [32] was used within the quasi-experimental design to support the redesign the referral management system and improve the patient environment at the study site clinic.
The methodology was applied using the DMAIC (Define, Measure, Analyze, Improve, Control) improvement framework [33] which consists of five distinct phases:
- Define the problem, identify required improvement activity, the opportunities for improvement, the project goals, and customer requirements.
- Measure process performance current process capabilities and identify the most important service parameters;
- Analyze the current process to determine root causes of unwanted variation and NVA (non-value add) activity;
- Improve process performance by piloting solutions to address and eliminate the root causes of NVA;
- Control the improved process and future process performance through measurement of key process performance metrics.
Where the Lean Six Sigma DMAIC framework has been utilized in healthcare settings, it has demonstrated successful outcomes [30, 34, 35] which have been categorized as delivering positive outcomes for the organization, patients, their families, and staff [23]. Additionally, Lean Six Sigma has been shown to streamline outpatient services [36].
2.3 Lean Six Sigma Toolkit
Within the Lean Six Sigma DMAIC improvement framework used for this study, we utilised appropriate Lean Six Sigma tools within the relevant phase/s of the DMAIC framework (table 1). The table classifies the Lean tools according to their purpose (table 1, column 2) and stage of use (table 1, column 3) within the DMAIC process. The effective deployment of these Lean Six Sigma tools contributed to the team’s ability to achieve their goals.
Table 1. Lean Six Sigma tools used in this study.
1.Improvement Tool
|
2.Purpose
|
3.DMAIC Phase Use
|
Project Charter
|
The main problem statement is defined in the project charter. It is used to identify the aims and scope of the project [37]
|
Define
|
SMART
|
SMART is used to manage those goals, the acronym stands for Specific, Measurable, Achievable, Relevant and Timebound (SMART) [38]
|
Define
|
SIPOC
|
A SIPOC (Supplier, Input, Process, Output and Customer) defines the customers and stakeholders and shows the process steps [39]
|
Define
|
VOC
|
The Voice of Customer (VOC) tool gathers customers feedback about their experiences [40]
|
Throughout all stages, iteratively seeking customer feedback
|
Gemba
|
Observation of the actual process taking place [41]
|
Define, Measure, Control
|
Ishikawa/Fishbone
|
Identifies root causes, representing the effect and the factors or causes influencing it [42]
|
Measure,Analyze,Improve
|
FMEA
|
Failure Mode and Effect Analysis is a risk analysis tool. It highlights the areas of process that require improvement [43]
|
Measure, Analyze, Improve,
Control
|
Process map
|
Process mapping increases understanding of complex systems [44]
|
Throughout all stages, from
initial development, current state to future state development.
|
TIMWOODS
|
A useful tool wherein each letter stands for one of eight potential wastes: Transport, Inventory, Motion, Waiting time, Over-processing, Overproduction, Defects and Skills [45]
|
Measure, Analyse, Improve
|
2.4 Data Gathering
A Project Charter [37] was developed by the Project team, discussed with staff working in the study site clinic, and the use of the SMART assessment [38] identified the following issues as key areas for data collection:
- Referral process (from writing of referral to its triage and apportionment date): this was facilitated by VOC, Process mapping and Gemba.
- System of tracking and triaging referrals: this was facilitated by VOC, Gemba, patient chart audit.
- Communication back to the referrer: this was facilitated by VOC, Gemba, records of mode of communication back to referral source.
- Time taken to register a referral; triage a referral; and book the patient's appointment: this was facilitated by VOC, Gemba, patient chart audit.
- Environment of clinic: this was facilitated by VOC and Gemba.
- Accessibility audit concerning seating and carpark signage: this was facilitated by audit, VOC, Gemba, and process mapping of current access and egress patterns.
Within the Person-centred Lean Six Sigma model [26], the Lean Six Sigma voice of the customer approach to understanding customers’ requirements has been shown to be synergistic with person-centred care practices, which utilize observations, narratives, conversations, focus groups and workshops [13]. The project team therefore understood and appreciated the value of spending time in authentic customer engagement through extensive voice of the customer sessions with the clinic staff and with patients.
2.5 Data relating to the referral process
Referral letters to paediatric ophthalmology services in Dublin come from a variety of sources (Table 2.) Any attempt to improve the referral process therefore had to account for all individual referral groups. To effectively implement and sustain change it is necessary to involve all stakeholders, and in the case of this study, extensive VOC engagement with the referrers to the paediatric ophthalmology service were key to the success of the improvement. VOC sessions with referrer groups sought to understand what, from a referrer’s point of view, worked well and what didn’t work well with the current referral system.
Table 2. Referral Sources.
REFERRAL SOURCE
|
TOTAL REFERRALS N=173
|
COMPLETE N = 173
|
School Screening
|
99
|
93
|
GP
|
27
|
4
|
Opticians
|
5
|
5
|
Tertiary Hospital
|
19
|
14
|
Public Health Nurse
|
22
|
1
|
Ophthalmologist
|
1
|
0
|
TOTAL
|
173
|
117
|
Following an audit of referral letters from referring sources (table 2), we were able to establish that they were often lacking the basic clinical information (Figure 2) required for the triaging of patients to categorize them based on the severity of their presenting complaint [46]. Triaging referral letters is an integral part of the outpatient clinic operation. Inadequate information in the referral letter adversely affects the triaging of the child which ultimately deems how quickly they are seen by the relevant clinician. When the project team began their improvement work, 33% of referrals were not right the first time. This meant that administrative staff had to repeat review the referral and triage details before they could schedule a child to the appropriate clinic and clinician.
Referrers, during VOC sessions were shown the audit data, and whilst recognizing the problems inherent in the current open nature of a referral letter, voiced concerns about any new referral form or system containing too many mandatory data sets and advised that these should be kept to an agreed minimum essential criteria. Referrers also asked that any new system should enable them to receive feedback confirming receipt of the referral and its current status so that they could update patients /carers.
2.6 Use of Gemba
Gemba is effectively a real-time observational study of a person in the place where the work or activity occurs, mirroring the Japanese concept of Kaizen (change for the better) [41]. This form of non-judgmental observational study is not unique to Lean, with the Person-centred Lean Six Sigma model highlighting the use of workplace observations in Person-centred work to measure and evaluate ‘where we are now ’ [47]. To support our data collection, we completed three Gemba walks, carried out at intervals by different members of the team to remove any potential of a singular viewpoint. From a child and family perspective, children were observed sliding off unsuitable seating designed for adults, and with no access to tables, completing their schoolwork on the clinic floor. Clinic administrative staff were observed spending an average of 20 minutes per triage letter gathering missing information. The Gemba gave the project team an overview of the clinic activity and assisted in the development of a process map (Figure 3). The process map was a product of collective intelligence and was validated by all stakeholders involved in the process. Process mapping is not simply about drawing the map itself but is also a process of building trust and consensus showing respect for those who carry out the actual work and empowering staff by involving them in co-designing any new process [44].
2.7 Consolidated data analysis
Gemba observations validated areas of NVA that had been highlighted by stakeholders within the clinic. The current patient management system was underpinned by the largely paper-based process which was human resource-heavy and according to the administrative team ‘not fit for purpose’. This equated to a staff confidence in the accuracy of the current referral management system of 70% but a satisfaction rating of 38% with the process workload. Within this paper-based process, during one Gemba, referral letters were noted as missing necessary patient information in 90% of cases. This was far above the average noted in the baseline data of 33% of referrals not being Right First Time (RFT), indicating potential for wide variation within the referral process, and enforcing the need for referrals to have minimum data requirements that were easy to monitor.
Administrative staff manually received referrals, triaged them and managed the clinics using this paper-based approach. The project team made use of the TIMWOODS tool [45] to further classify the NVA identified by the VOC, Gemba and process maps.
Table 3 gives an overview of NVA that patients, their families, administrative staff and clinicians experienced due to the existing processes within the clinics.
Table 3. TIMWOODS analysis.
|
Waste
|
Impact on child and family
|
Impact on staff
|
T
|
Transport
|
Delay in arrival due to poor directional signage to clinic
|
Transporting paperwork Excessive movement of staff, forms/patient files between rooms.
|
I
|
Inventory
|
Adult orientated furniture and fittings
|
Paper based referral process
|
M
|
Motion
|
Lack of motion due to clinic flow
|
Excessive motion gathering paperwork
|
W
|
Waiting Times
|
Waiting times for appointment
Waiting time within clinic
|
Waiting time for queries to referrers to be answered
|
O
|
Over-Production
|
Unclear about appointment and its purpose-leading to multiple non attendances
|
Follow up queries on referrals
|
O
|
Over- Processing
|
Appointments rescheduled
|
Adding pre-check phone calls for all patients to avoid DNA
|
D
|
Defects
|
Poor environment of care
|
Missing information in referral forms
|
S
|
Skills
|
|
Clinicians follow up with referral sources
|
Further discussion of our TIMWOODS analysis with stakeholders revealed the following:
- Gemba observations evidenced up to four staff involved in the processing of one standard referral received. There was excessive movement of staff and patient files between rooms.
- An absence of any referral management system produced excessive motion for administrative staff.
- Patient flow within the clinic was heavily influenced by administrative staff processing arriving referrals.
- Administrative staff were observed contacting patients' parents/families in advance to mitigate against DNA.
- Clinical staff following up with referrers and patients were missed appointments had occurred.
- Administrative staff spending time on processing referrals and offsetting DNAs, and not developing solutions.
When staff were unable to contact patients' parents to follow up on referrals via SMS (text message) or phone, it often resulted in non-attendance (Do Not Attend/DNA) which averaged at 20% per clinic. This increased the time children were waiting to be seen by the appropriate clinician. The time spent by administrative staff trying to contact these parents had a corresponding impact on clinic flow with administrative staff diverted to follow up calls increasing waiting times for children and their families in terms of their actual waiting time in the clinic on the day of their appointments.
VOC sessions with the children, their parents, and staff showed consensus that the clinic environment itself was not fit for purpose and was not designed with the child in mind. Families interviewed were struggling to find the clinic, consequently arriving late to their appointments. There was poor seating, no stroller parking, limited signage, poor contrast sensitivity and a poorly illuminated environment. All of these issues impacted on visually impaired children.
These themes representing NVA were then further evaluated with clinic staff in a further series of workshops facilitated by the project team, using an Ishikawa diagram (Figure 3) enabling identification of potential root causes of the issues to be considered [42] to facilitate solution generation within the clinic.
After brainstorming solutions with the clinic staff, a second modified Ishikawa Diagram was developed in collaboration with them (Figure 4). The potential highest impact solutions to the root causes identified were the implementation of a standardized referral management system, an online patient appointment system, in addition to a review of the current physical layout of the clinic to enhance the environment of care.
Following the development of the Ishikawa diagrams, the team made use of a Failure Modes and Effects Analysis (FMEA), which is a systematic method to identify where and how a process might fail and to assess the relative impact of different failures, in order to identify the parts of the process that are most in need of change [43].
Working with staff, building on our collected data and our process maps, we identified potential failure modes for each of the co-designed solutions of an online referral form, online patient management system and child-friendly environment, identified within the Ishikawa exercise (figure 4).
Potential failure modes for an online referral form were:
- incomplete referral forms and referral duplication as well as referrals sent to the wrong person and not being triaged on time.
Potential failure modes for an online management system were:
- software malfunction
- amending templates which required external support
- inadequate staff training resulting in an inability to operate the system
Potential failure modes for implementation of the child-friendly environment were:
- potential budgeting issues and problems with procuring stock.
The use of the FMEA in conjunction with the use of the Ishikawa tool (figures 3-4) enabled us to collaborate with and codesign solutions with the staff working within the clinic.
2.6 Improvement/solutions
The co-designed solutions focused on the main failure modes as outlined below.
2.6.1 Incomplete referral forms
In order to improve the statistics on right-first-time referrals, both staff at the study site, and referrers agreed that a move to an electronic online referral form was the best option. This would enable patient status check at a glance and comprise compulsory fields to obviate NVA in follow up activity for administrative staff. The following fields were considered the minimum compulsory data required for any new online referral form:
- Name DOB, address, gender
- Visual acuity measurement
- Date referral
- Name of referrer
- Additional needs
2.6.2 Electronic referral system
Staff reached consensus that the referral management system process should move from a manual paper-based system, and proposed using the project findings to support a business case to implement a new electronic system and move to electronic online referrals.
2.6.3 Implementation of a child-friendly environment.
It was agreed there was a need to procure appropriate children's chairs and tables with associated accessories (coloring books and crayons) for the waiting area (figure 5). In addition to this, we conducted an access improvement audit with the help of design specialists already working within the ophthalmology service to generate recommendations for signage and colors suitable for visually impaired children.
Following the outcomes of the access audit, we implemented a Wayfinding Strategy, including Assistive Technologies to positively impact the safety and psychological wellbeing of clinic attendees. This took the form of the provision of new accessible signage, inclusive of patient and family approach to the clinic, the lift/elevator access and a means to easily locate the Eye Clinic reception to ensure it is accessible to everyone. We also ensured that all pre-visit correspondence and information relating to access and wayfinding was consistent with on-site directional information. Upon review of internal environments, it was noted that backgrounds, color contrast and lighting were completely inadequate for people with visual impairment. To ensure safety during navigation we worked to ensure legibility of signage and adequate illuminance throughout.
2.6.4 Funding the improvement
The problem statement, collected data, findings of the root cause analysis and potential solutions generated by the staff working within the clinic were presented to the ophthalmology service management team. The average time spent on incomplete referrals at 20 minutes, and the 33% of referrals that were not right first time, alongside staff, referrer and patient voice of the customer data were seen as a compelling case for change.
Funding for the online referral system was successfully secured from the annual allocated Health Services Executive budget for the locality. The ophthalmology service was familiar with online providers of a scheduled booking system, that was currently in use in phlebotomy services, and were able to leverage existing agreements to secure an extension of the online scheduled booking system to the ophthalmic clinic. Furniture suitable for the demographic of a paediatric clinic was secured from existing stock within the wider service, and the access and wayfinding strategies were cost neutral as part of an ongoing review of signage overall.