Article Text

Restructuring of the Diabetes Day Centre: a pilot lean project in a tertiary referral centre in the West of Ireland
  1. A M McDermott1,
  2. P Kidd2,
  3. M Gately3,
  4. R Casey3,
  5. H Burke3,
  6. P O'Donnell4,
  7. F Kirrane5,
  8. S F Dinneen3,6,
  9. T O'Brien3,6
  1. 1Discipline of Surgery, National University of Ireland, Galway, Ireland
  2. 2Department of Pharmacy, Galway Roscommon University Hospital Group, Galway, Ireland
  3. 3Department of Diabetes and Endocrinology, Galway Roscommon University Hospital Group, Galway, Ireland
  4. 4Leading Edge Group, Galway, Ireland
  5. 5Department of Medical Physics, Galway Roscommon University Hospital Group, Galway, Ireland
  6. 6Discipline of Medicine, National University of Ireland, Galway, Ireland
  1. Correspondence to Dr Ailbhe M McDermott, Discipline of Surgery, School of Medicine, Clinical Science Institute, National University of Ireland, Galway, Ireland; ailbhemcdermott{at}gmail.com, www.nuigalway.ie/surgery/research

Abstract

Introduction Diabetes is a chronic disease amenable to management in the community and outpatient setting. The increasing incidence of diabetes places outpatient endocrinology services under pressure to provide a quality service in a timely manner. Our aim was to apply lean thinking to the diabetes clinic in a tertiary referral centre in the West of Ireland to improve flow, as reflected in reduced patient journey times.

Methods The project lasted 6 months, from January to June 2011. An introductory seminar on lean thinking was arranged to inform and motivate the Diabetes Day Centre staff. Two ‘rapid improvement events’ took place. Value stream mapping (VSM) was the predominant lean tool employed. Patient journeys were mapped and quantified (minutes) using timesheets allocated to each step in the process at baseline, and following intervention. Data were analysed using Minitab V.16.0.

Results VSM allowed the value-adding and problem-causing steps in the patient journey through the diabetes clinic process to be identified and addressed. Total patient journey time through the clinic was significantly reduced from 118 (±38.02) min to 58 (±18.30) min (p<0.001).

Conclusions This project reflects the successful application of VSM as a lean tool in a pilot study at our institution as evidenced by improved patient flow and a significant reduction in patient journey time through the clinic. Through the incorporation of Lean into the ethos of the hospital, we have the potential to deliver excellent care in a safe environment and in an efficient manner, while benefiting the patient, employees and tax-payer.

  • Lean management
  • Process mapping
  • Patient-centred care
  • Diabetes mellitus
View Full Text

Statistics from Altmetric.com

Introduction

The prevalence of diabetes mellitus is continuing to rise, with reports of rates having reached ‘epidemic’ and ‘pandemic’ proportions.1 ,2 Diabetes mellitus is a chronic disease that is largely amenable to management in the community setting.3 However, in Ireland and other countries alike, continuing care of patients with diabetes mellitus is rarely provided solely in general practice. Consequently, the demand being placed on outpatient (ambulatory) services to provide high quality care for an exponentially increasing patient population is approaching an unsustainable level in several institutions in both in Ireland and worldwide. This is evidenced by increasing waiting list times and overcrowded outpatient departments.

In recent times, healthcare organisations worldwide have adopted ‘lean principles’. Lean is a quality improvement philosophy which in essence strives to improve value, as defined by the customer (patient), and remove waste, resulting in more efficient practices in the continuous pursuit of perfection. It was originally conceived for manufacturing by Ohno Taichi of Toyota, as the Toyota Production System (TPS) and became known as ‘Lean’ in the 1990s.4 ,5 Restructuring the delivery of patient services by implementing lean principles provides several benefits. Economic benefits can be achieved by improving patient flow, increasing efficiency and reducing administration and staffing requirements.6 ,7 Patient (customer) satisfaction is increased as a result of improved service delivery, less waiting time and improved quality of care, while an increased sense of empowerment and ownership enhance the morale of staff, who provide an improved quality of care with a more desirable use of their expertise.8 The clinical relevance of understanding value from the customer (patient) perspective is reinforced by the established link between patient compliance with treatment and patient expectations and preferences, which ultimately contributes to improved outcomes.9–11

The outpatient process in most institutions is composed of a series of defined steps suggesting the feasibility of applying lean principles to process restructuring in an effort to improve efficiency. Value stream mapping (VSM) is a lean tool that can be used to plot and analyse the individual patient journey in this setting. A visual representation of patient movement throughout the system can aid the identification of the useful (value-adding) steps and waste (non-value-adding) steps. It also facilitates recognition of bottlenecks, handoffs and prolonged waiting times. The patient journey can then be redesigned to improve quality, safety, service and cost in a patient-centred approach.

The aim of this study was to utilise lean principles to restructure a diabetes clinic as a pilot lean project at our institution, Galway Roscommon University Hospital Group (GRUHG), Ireland, in an effort to improve patient flow through the clinic, quantified by patient journey (throughput) times.

Methods

Setting and design

The study took place in the Diabetes Day Centre, an outpatient unit in University Hospital Galway part of the GRUHG, a tertiary referral centre in the West of Ireland with over 67 000 outpatient consultations annually. Patients are referred to the Diabetes Day Centre by their general practitioner. There are a number of steps each patient must pass through after arriving at the clinic as diabetes is a chronic illness and management involves a multidisciplinary approach with input from allied health professionals.

Intervention

Project duration was set at 6 months. The core team consisted of employees with a range of roles across the service: two consultant endocrinologists, the administrative manager of the Diabetes Day Centre, a clinical nurse practitioner in diabetes, a non-consultant hospital doctor and two hospital employees with an interest in the implementation of lean principles. There were no formalised roles or project champions within the team.

VSM was chosen as the most appropriate lean tool for this pilot lean project as it is an excellent diagnostic tool for identifying problem areas in a short duration of time. It can also facilitate the selection of other appropriate tools for use.

A literature review was performed to determine patient preferences and factors influencing satisfaction (eg, value-adding) with respect to outpatient clinic visits. From staff experience and as evidenced in the literature,12–14 total time spent in clinic and time spent waiting to see a physician were major factors influencing patient satisfaction and were chosen as the key variables in this study. A key objective of the redesign, therefore, was to improve the flow of patients through the clinic to reduce the overall patient journey time.

Staff education: creating the need for Lean

Educating employees of the diabetes centre was important to ensure understanding and facilitate cooperation with the initiative. This took place during a weekly lunchtime team meeting. The quality improvement initiative was termed a ‘rapid improvement project’.

Current state mapping

The VSM exercise took place during a ‘rapid improvement’ or ‘Kaizen’ workshop at a morning diabetes review clinic. This workshop had two goals: to gather baseline data on patient flow through the clinic and second, to gather information about each step in the process. Patient journeys (n=41 before, n=32 after) were mapped using timesheets located at each step throughout the process (see online supplementary data). Start time was defined as the time when the patient presented to reception. The end point was when the patient left the outpatient department. It was stressed that the important feature was not the value-added time spent with the patient, but rather the wait time (waste) during which the patient was in the waiting room. As it would not be practical to remove employees from clinical and administrative areas for a whole day or at the same time, it was decided that staff members would take part on an individual basis. This method is beneficial as it does not interrupt the flow of the clinic, permits data to be collected on patient flow through the clinic, and is an inclusive exercise. A fundamental concept in lean methodology is that the people best qualified to analyse and improve a workflow are those actively involved in the process on a daily basis. A group debriefing session including all employees took place during lunch after the clinic. Each step in the process was discussed and deemed as waste (non-value add), value-add or necessary-non-value-add. In brief, any step which contributes towards management of the patient's diabetes mellitus was deemed to be value adding. A current state map was constructed (figure 1). This permitted the complexity of the patient flow through the clinic to be visualised. It also revealed areas of poor quality staff–staff interactions, unnecessary travelling within the department, re-work and bottlenecks.

The technique of VSM requires that the team consider the ‘ideal state’, so that changes can be made in an attempt to bridge the gap between the current and ideal state. As a group, it was decided that the ‘ideal state’ would be comprised of the following features:

  • Patients would arrive at the clinic just before their appointment time.

  • Patient arrival times would be staggered according to the different patient types/streams.

  • There would be clear visual targets (key performance indicators) for each staff member.

  • The patient journey throughout the clinic would be easy to follow for the patient, improving patient expectation.

  • If any wait-time did occur, it would be used as an opportunity to provide education.

  • Patients would move throughout the steps in the clinic with no delays (single piece flow).

  • There would be no re-work.

  • There would be a full compliment of staff at each clinic arriving on time, and no shortages of staff.

  • Recent HbA1c levels would be available for all patients and could be performed as a point of care test if necessary.

  • Total journey time would only include the value-adding activities.

The group then constructed a future state value stream map (figure 2). From experience, the group generated a list of problems and potential solutions, from which three were chosen for implementation (action plan) as part of this project. An ideal target was set: a reduction in patient journey time of 50%. To implement and sustain the initiative, the projects were discussed weekly at the team meeting. A re-evaluation rapid improvement event took place after 4 months.

Figure 1

The current state map was constructed after value stream mapping took place during the initial rapid improvement event. The major clinic steps encountered by all patients are coloured in blue. Only some patient types (annual review, new diagnosis, etc) must pass through the other additional steps (orange). The flow of information recorded in the DIAMOND software program, patient files which accompany the patient and PAS (the hospital laboratory result system) is complicated and haphazard. C/T, cycle time.

Outcome metrics

The focus of this study was patient journey time. During each rapid improvement event, individual task owners at each step in the process were responsible for logging the name of each patient, the start and finish time of each step, the availability of HbA1c results (physician only) and where the patient was directed to next (see online supplementary data). This activity was performed at an annual review clinic on both occasions (before and after the intervention), with a similar case-mix and patient number. Metrics evaluated as part of this study were as follows:

  • Mean patient journey time, defined as time from registration at clinic reception to the recorded time the patient physically left the department;

  • Mean time to see a physician, defined as the time from registration to the recorded time of physician encounter (ie, door to doctor time).

Statistical analysis

Statistical analysis was performed using Minitab V.16.0 (Minitab, Coventry, UK).The Kolmogorov–Smirnov test for normality was conducted. Differences between the traditional (baseline) and reformed (after) pathway in mean patient journey time and mean time to see a physician were determined using the t-test or Mann–Whitney U test as appropriate. p Values less than 0.05 were deemed to be significant.

Results

Nine value-adding steps in the diabetes outpatient process were identified (table 1). Seven of these steps directly contributed towards management of the patient's disease. Two of these steps are necessary for every patient attending the clinic, while the remaining five are required by individual patients according to their appointment type and patient stream (new patient, annual review, well controlled, etc).

Table 1

The value-added steps of the patient journey through the diabetic clinic

The current state patient journey (value stream map, figure 1) was constructed by the team during the debriefing session of rapid improvement event 1 using a large sheet of paper and a marker. The journey of 41 patients was tracked during this first rapid improvement event. Mean baseline total journey time (lead time) was 118.14 (±38.02) min, ranging from 35 to 195 min. Mean total time to see a physician was 61.26 (±33.08) min, ranging from to 15 to 140 min. The problem list generated during rapid improvement event 1 is outlined in table 2.

Table 2

Problem areas and potential solutions in the Diabetes Day Centre

Figure 2

The future state map shows a realistic goal to bridge the gap between the current state and the ideal state. The major clinic steps in blue are encountered by all patients. Only some patient types (annual review, new diagnosis, etc) must pass through the other additional steps (orange). C/T, cycle time.

The specific changes implemented in restructuring the diabetes clinic were as follows:

  1. Improved throughput by addressing the bottleneck caused at the nurse's station.

Traditionally, patients follow the steps in the order outlined in table 1 with the nurse's station preceding the physician consultation. During this session, the nurse records the patient's weight, height and blood pressure and reviews the patient's glycaemic control and insulin regimen (or oral hypoglycaemic agents, as appropriate). These data are recorded on DIAMOND (a diabetes clinic software program) so that they are available for the doctor. There are three nurse stations, and frequently up to six doctors, with the result that doctors often wait on patients, despite a number of patients being in the waiting room. In order to overcome this bottleneck, it was agreed that in these circumstances, the doctors would see the patients who are in the waiting room first, and record their blood pressure, weight and height themselves as necessary. Any patient who requires further education or a discussion regarding diabetes management can then visit the nurse if required.

  • Work standardisation by introducing a standard method of referral for retinal screening

There was no standard method of referring patients for retinal screening, an optional step for some patients in the process. Retinal screening is required on an annual basis in the management of diabetes mellitus. Some patients were referred directly by the nurses, potentially missing their physician consultation slot, while others were referred by the physician. However, the referral method was inconsistent. Some doctors telephoned to inform the retinal screening technician a patient needed to be seen, while others physically walked with the patient, but if the screening room was occupied by another patient they guided the patient to the general waiting room, leaving the chart outside the door with no further instructions. This heterogeneous referral pattern meant that patients were often not assessed in the order in which they were technically referred. In some instances during our baseline analysis, patients were left sitting in the waiting room for so long that they prematurely left the clinic. A single standardised method of referral was implemented. The referrer, preferably the physician, phoned the retinal screening technician. If the technician was available, the patient was walked down and assessed immediately. If the technician was busy, the patient’s name was taken and the patient was shown back to the waiting room.

  • Standardised clinic start times

Different consultants started their outpatient clinics at different times, ranging from 8:00 to 9:30. Patients were always onsite from 8:00 onwards (avoiding morning traffic, parking difficulties, etc). Frequently, morning clinics would run late, thus delaying the afternoon clinic. As a direct consequence, doctors in particular would have to stay late (resulting in overtime costs) in order to complete their clinic and ward responsibilities. This contributes towards non-consultant hospital doctors (NCHDs) working in excess of the 48 h limit specified in the European Working Time Directive. It was decided that all clinics would commence at 8:00. If ward rounds were being conducted, the team could select one or two physicians to start seeing patients with the full complement joining once the ward round was completed. Alternatively, the ward round could take place after the clinic.

Re-measurement: any improvement from baseline?

The new patient pathway was implemented and piloted for 2 months before re-analysis took place during the second rapid improvement event. Data were collected on 32 patients in this re-analysis exercise. Total patient journey time had decreased significantly (p<0.001, t test) from 118.13 min to 58.15 (±18.30) min, ranging from 31 to 99 min. This equates to a 49.2% reduction in patient journey time, almost reaching the target of 50%. An overview of key before and after metrics is presented in table 3.

Table 3

Key before and after metrics

Discussion

The evidence in favour of the lean approach to healthcare is rapidly expanding. However, Irish hospitals in both the public and private sectors have been slow to adopt this methodology. In the current economic environment, hospitals throughout the country are trying to improve patient-level services while coping with increasing financial constraints and prolonged waiting lists for in- and out-patient services.

This pilot study reflects the promising potential for lean principles in the Irish healthcare sector. Simple but effective changes were identified from within the workforce, by the employees of the diabetes clinic, and implemented in a cost neutral manner. Total patient journey time was significantly reduced by 49% from 118 to 58 min as a result of a 60 min reduction in waiting time (waste). Interestingly, there was also a significant reduction in the time spent in consultation with the doctor, which decreased from 26 to 18 min. This is surprising considering that some patients bypassed the bottleneck at the nurses’ station. A plausible hypothesis to explain this reduction in consultation time could be that less time was spent by the NCHDs waiting to discuss cases with the consultant. Perhaps the VSM exercise created an increased awareness of patient journey time, and the knock-on effects of delays within the process.

VSM, the main lean tool chosen for execution of the project, played a pivotal role in achieving this success. VSM is the preferred tool to implement a lean approach.15 It can be performed in a timely manner, is inclusive as all task owners are involved and ensures the benefits accrue to the whole process and not just isolated islands in the process. VSM allows each member of the team to understand where he or she fits into each process. However, this is largely dependent on upfront communication; it needs to be emphasised that the intervention is not a threat to employment. This lean tool highlights the process or stream nature of the clinic, where the actions of one person can have repercussions for patients and colleagues up- and down-stream. The challenge lay in determining how to remove the waste and address the bottlenecks in a cost neutral manner. VSM facilitates the identification of the role of other lean tools in the restructuring process.

Other studies have successfully implemented lean changes in an outpatient setting, although few published studies specifically pertain to the diabetes setting. A lean transformation of an out-patient department  phlebotomy service in a teaching affiliate to Harvard Medical School reduced the average patient waiting times from 21 to 5 min, and improved patient satisfaction.16 Similar benefits from lean improvement strategies in the outpatient environment have been demonstrated in the UK and the Netherlands.17 ,18 There are indeed other benefits, in addition to workflow improvement, from lean thinking service redesign, although these were beyond the scope of this pilot project. Improved patient satisfaction is often associated with a reduction in non-attenders, and improved patient adherence. This is frequently an issue in chronic diseases such as diabetes mellitus. Another significant advantage to the lean approach to change management is the requirement for its endorsement by and involvement of staff. Employees often oppose and resist change, which is often reflected by lack of interest and reluctance to adopt new roles, thus jeopardising the sustainability of service redesign. However, the lean approach encourages staff involvement, recognising that process owners are the experts of their step in the process. This improves engagement and enthusiasm, as was the case in this project, with a substantial improvement in employee morale.

There are recognised limitations to the study. Lean philosophy is such that ‘value’ should be considered from the patient's perspective. However, as this was the first encounter by diabetes clinic staff with the implementation of lean principles, we did not include a patient on the core team, or record any patient-centred data or questionnaire analysis of the improvement strategy. We considered patient value as perceived from the employee's point of view. Future studies will include a patient-centred data capturing method to provide insight into what is truly important at the patient level. Another significant limitation to the study was the lack of financial support. This project was completed in a cost neutral manner, which restricted the improvement strategies that could be implemented. Similarly, it required employees to sacrifice lunch breaks and personal time to take part in group discussions. Additionally, the data captured as part of this project focused on patient journey time, as this was the fundamental focus . It was presumed that once the patient was seen, the quality of the service provided was equal for all patients. We did not determine if any implemented improvement impacted on other areas such as patient management, safety or employee morale, for example. The benefits of lean principles in this regard are well described and accepted.

This was a pilot lean project in an organisation with no previous experience of the implementation of lean principles. The aim was to reduce patient journey times through the clinic and to that end it was successful. This initial study was performed at a high level. The next move is to focus on each individual separately and consider individual tasks within the steps, so that further time can be saved by preventing duplication of work. If these changes are to be viable in the long-term setting, lean thinking needs to be incorporated into the ethos of the diabetic clinic, and GRUHG at an organisational level. The future implementation of lean principles in the diabetes clinic is looking bright, with numerous follow-on projects currently underway, in the spirit of continuous improvement. The support of GRUHG senior management and the consultant endocrinologists was crucial to empower the front-line staff to engage in continuous improvement strategies, and to recognise and celebrate the success of the implemented improvements.

Conclusion

Restructuring the diabetes outpatient clinic using lean principles was successful in reducing patient journey time from 118 to 58 min (p<0.001), reflecting the successful application of VSM as a lean tool in this pilot study in GRUHG. The application of additional lean tools could potentially further restructure the diabetes clinic to enhance the patient journey. The lean culture for improvement in continuous pursuit of excellence has also expanded to other departments in an attempt to improve the quality of service delivery and patient safety at an organisational level. Through the incorporation of Lean into the ethos of the hospital, GRUHG can deliver excellent care in a safe environment and in an efficient manner, while benefiting the patient, employees and tax-payer.

Acknowledgments

The authors would like to thank the patients and staff of the Diabetes Day Centre, Galway Roscommon University Hospital Group.

References

View Abstract

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:

Footnotes

  • Contributors All authors listed above meet the requirements for authorship. All authors (AMD, PK, MG, RC, HB, POD, FK, SFD and TOB) contributed substantially to conception and design, acquisition of data or analysis and interpretation of data. All authors contributed towards drafting or revising the manuscript and all authors approved the final version of the manuscript.

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.