Article Text
PDF
Abstract
Objectives This article describes key changes made within one hospital pharmacy department in the UK during on-going work to improve services and meet the aims of the National Health Service (NHS) Quality, Innovation, Productivity and Prevention (QIPP) agenda. A series of projects designed to significantly increase the value of the service; reduce waste, increase efficiency and improve clinical services, without employing more staff, were undertaken.
Methods The department used various change management techniques and lean methodologies to bring about change. This has been reinforced by robust and continual data collection and interpretation via run charts, to objectively demonstrate improvement.
Results Average prescription turn-around times are approximately 25 minutes each day or less. Delays in provision of chemotherapy have fallen from around 60% to less than 5% of patients each day. Clinical pharmacists see approximately 98% of available inpatients each day and on average 92% of patients have their medicines reconciled within 24 h of admission. The Trust Medicines Management Improvement Group has also run a number of successful improvement projects to improve medicines safety.
Conclusions The use of common change management techniques are applicable to hospital pharmacy and can result in measurable improvement in service delivery. It also indicates that the small size of a department is not necessarily a handicap to achieving better quality and efficiency in dispensary and chemotherapy production units, or in measurably addressing local and national patient safety concerns, without the need for additional staffing resources.
- Clinical Pharmacy
- Drug Storage And Distribution
- Pharmacy Management (Organisation, Financial)
- Pharmacy Management (Personnel)
- Quality Management
Statistics from Altmetric.com
- Clinical Pharmacy
- Drug Storage And Distribution
- Pharmacy Management (Organisation, Financial)
- Pharmacy Management (Personnel)
- Quality Management
Introduction
This article describes key changes made within one hospital pharmacy department in the UK during ongoing work aimed at improving the quality and efficiency of its pharmaceutical services. The changes were implemented in response to various factors and over several years have significantly increased the value of the service; waste has been reduced, efficiency increased and clinical services improved, without employing more staff. The target audience for this article is anyone wanting to improve service efficiency and increase output without the need to invest significantly in additional resources. This paper is not a review of improvement methodologies1–3; rather it outlines how the application of these techniques has transformed a service.
Musgrove Park Hospital is an acute district hospital with 621 beds, admitting around 85 000 patients annually and generating around 99 000 Finished Consultant Episodes (FCE) (2012). In 2013, the pharmacy employed 49.4 wholetime-equivalent staff. When staffing levels were standardised against hospital workload (ie, staff/1000 FCEs or 1000 admissions) it ranks as relatively small (see figure 1).
South West Area HA hospital pharmacy staff 2012–2013.
Historically, the department had operated a range of hospital pharmacy systems not dissimilar to those found in other UK hospitals. Systems and processes were reviewed and changed infrequently, and such reviews would often coincide with the arrival of a new staff member from another hospital. The department had no formalised method for reviewing and changing practices, and this, coupled with relatively low staffing levels resulted in frequent criticism of service quality. Complaints focussed on long waiting times for dispensed medicines, ward stock and chemotherapy. Staff spent much time and effort attempting to resolve the same range of acute, operational problems each day, often describing this as ‘fire fighting’. However, little effort was spent exploring permanent remedial actions, and so these operational problems continued to occur. This had negative consequences for service quality in other areas, one example being the provision of ward-based clinical pharmacy which was fragmented, and its sustainability sensitive to staffing levels and daily dispensary workloads.
The situation was further complicated by numerous, long-standing service-level agreements that required the department to provide services to external organisations. These commitments were often given precedence over service provision to hospital inpatients. This was illustrated by the fact that dispensary operations were designed to match the departure schedule for transport vehicles delivering medicines to external organisations. Only when medicines were made ready for transport would dispensing begin for acute hospital patients. Delays were effectively ‘built in’ to the system.
In 2010, the department of health published a document entitled ‘the NHS quality, innovation, productivity and prevention challenge: an introduction for clinicians’4 (QIPP). It focussed on the concept that high-quality healthcare in the UK would in the near future be delivered in a reliable and cost-effective fashion. The expectation was that service delivery would require redesign and there would be an emphasis on error prevention (process unreliability), efficiency and productivity. The document and subsequent related publications promoted the use of a wide range of change management techniques to achieve this end. These techniques had been used in various forms for decades in manufacturing and IT industries, and the intention was that they could be used to reconfigure health services to attain better outcomes. Subsequently, 12 formal, national work streams were set up with the expectation that the NHS as a whole would embrace this new approach to improvement. The work streams are as follows, and illustrate the range of activities believed to be amenable to the QIPP approach:
-
End-of-life care; where higher numbers of patients would be identified as being close to the end of life, and where services would be redesigned to ensure they lived and died well in the location of their choice.
-
Effective management of long-term conditions. This was identified as being important as around 70% of all health and social care costs can be attributed to this.
-
Effective medicines procurement and use.
-
Mobilisation. This involved taking a values-driven approach to service transformation.
-
Better procurement with better management of NHS non-pay spend.
-
Productive care delivered in a productive environment.
-
Right care delivered in a cost-effective way with maximised value.
-
Right shared care decision making; where it would become normal to make shared decisions rather than unilateral ones.
-
Safe care which focuses on harm reduction.
-
Workforce optimisation.
-
Urgent and emergency care delivery.
-
Primary care commissioning which focused on increased efficiency of care delivery and general practitioner operations, embedding of QIPP in primary care, and integration with other QIPP work streams.
A large amount of web-based information on the use of change management techniques was published to support QIPP work. Several of these techniques proved effective in helping to bring about desired changes in our organisation. This national quality programme offered our department an opportunity to publicise some of its efficiency work.5–7
Now several years on, service provision has improved significantly. Average prescription turn-around times are approximately 25 min each day or less (figure 2).8 Delays in provision of chemotherapy have fallen from around 60% to less than 5% of patients each day.9 Clinical pharmacists see approximately 98% of available inpatients each day (figure 3), and on average, 92% of patients have their medicines reconciled within 24 h of admission (figure 4).10
Daily prescription turn-around time and dispensing workload.
Proportion of inpatients reviewed by clinical pharmacists.
Proportion of patients with medicines reconciliation completed.
Key changes within the department
The following is a chronological list of key changes that were made in the department. These built upon one another, and the success and sustainability of one change supported the success of subsequent changes.
Photocopying the Medicines Administration Chart to use as a prescription for dispensing
In 2002, a business case was made to recruit eight medicines management technicians in order to implement a ‘dispensing for discharge’ system. Also known as ‘one stop dispensing’11 the system was and is considered to be the preferred model for handling dispensed medicines in UK hospitals. The business case was based on models in other local hospitals, and was rejected by the trust. It was therefore necessary to think differently, and after some discussions with enthusiastic ward managers a system for nurse-led drug ordering began to emerge.7
This simple system involved photocopying the Medicines Administration Chart (MAR), highlighting the required drug/s and sending the photocopy to the dispensary. Full implementation took over 3 years but the system had several advantages including
-
Ward staff could order the required drugs as soon as the need was identified. This was much more responsive, compared to visiting pharmacy staff transcribing orders and prescriptions. It significantly reduced the batching of work and enabled a lean pull12 system to be introduced to the dispensary. ‘Lean pull’ refers to the provision of dispensed medicines in response to an actual demand from the customer as opposed to anticipating requirements by forecasting.
-
Photocopying eliminated the wasteful process of transcribing and the errors associated with it.
-
Photocopier leasing was and is inexpensive, and the copying process quick and reliable. There was no need to invest in staff to transcribe orders for pharmacy to process.
-
The MAR remained available on the ward at all times.
-
Prescriptions written for discharge became redundant as the medicines required for patient discharge were already on the ward.
-
The system allowed ward stock levels to be reduced, even on the intensive therapy and high dependency unit.
To facilitate this process and eliminate the need to photocopy multiple prescription charts, the Trust MAR chart was redesigned so that patient demographics, allergy status and all prescription items could be documented safely on a single page.
Data collection and key performance indicators
In 2004, a new clinical pharmacy manager arrived, and with him, a new discipline for audit and data collection. A simple and effective tool for collecting and analysing the ward-based activities of all clinical pharmacists was introduced. The result was that in addition to amassing large amounts of robust data regarding pharmacist daily activities, the percentage of patients seen by clinical pharmacists rose from around 40% to 85% within 6 months of implementation. Each pharmacist became accustomed to being held responsible for the correct and timely collection of activity data. Data capture became mandatory. This marked the beginning of a move towards a data-centric departmental culture.
Introduction of the dispensing robot
In 2006, the hospital invested in an automated dispensing robot and although this improved performance in terms of accurate dispensing and stock control, average prescription turn-around times remained in excess of 3 h from prescription receipt to the time when ready for collection. However, the robot did reduce wasted motion as the dispensary ergonomic layout was reconfigured to reduce travel time as part of this project.
Working with the Institute for Healthcare Improvement (IHI)
In late 2006, the trust joined the Institute for Healthcare Improvement (IHI) Safer Patient Initiative (2).13 Over four years, the organisation and pharmacy department were introduced to the use of various, effective change management techniques during efforts to increase reliability of a range of trust-wide clinical processes. This reinforced the value for robust and continual data collection and its interpretation via run charts and statistical process control charts. Change was brought about through formal use of techniques, such as: process mapping, failure mode effect analysis, and the IHI model for change.1
When the project ended, some hospital clinical groups continued their innovative patient safety work using the same techniques. One of these included the Trust Medicines Management Improvement Group (MMIG) which had originally worked on medicines reconciliation rates (figure 4)10 and warfarin-associated adverse events (figure 5). Table 1 contains a list of successful MMIG improvement projects. It is likely that these projects, and disciplined adherence to new standardised processes have contributed to reduced pharmacy workload, although a cause and effect relationship has not been proven.
MMIG improvement projects
Warfarin-related adverse drug even rate.
‘Lean’ dispensary and chemotherapy production
In 2008, the hospital trust organised a 5-day course for managers on ‘lean’ systems design and patient flow. These techniques were subsequently used to markedly increase the efficiency of the dispensary and chemotherapy production unit. Dispensary average daily prescription turn-around times fell rapidly from over 3 h to 25 min8 (figure 2), and daily waits for chemotherapy from the production unit fell from >60% to <5%.9
The ‘leaning’ of these activities began with process mapping exercises. From this point onwards all improvement work was approached formally using a combination of lean and IHI change methodology techniques (see Improvement methodology).
Reviewing the profitability and value of external contracts
The trust had historically provided pharmaceutical services to local community hospitals and a mental health trust at below market value. Following negotiation and review it was decided that service provision would cease. Staffing levels in areas of the department where workload subsequently decreased were reduced through natural wastage. This change had a hugely positive impact on the flow of work to hospital patients, as they quickly became the sole focus of pharmacy activity.
Improvement methodology
From 2008 each potential improvement project was approached using the following steps:
Choosing appropriate metrics
It is essential to be able to monitor the effects of change during any improvement project. Without the ability to do this change cannot be objectively demonstrated, for example, when attempting to improve the workflow through a dispensary, appropriate measures would include prescription turn-around times, prescription takt (ie, the reciprocal of rate of work or the time it takes to do a piece of work) times and also dispensing error rates, included as a counter measure.
The metrics would be collected and recorded as a run chart.14 As time progressed, the dataset could be analysed using simple statistical process control techniques to determine whether changes were statistically significant over time.15 The figures contained within this article represent a variety of run charts plotting changes in specific measurements over time.
Variations within the data are to be expected, but outlying data points may warrant further analysis (eg, figure 4, Data points for June and July 2013). Variation may be random or ‘special cause’, for example, caused by acute staffing shortages, new staff being unfamiliar with processes, or current staff developing ‘workarounds’ to the correct process. Basic statistical process control analysis14 ,15 provides a means of objectively determining whether the data represented statistically significant change or normal random variation of the process. Timely analysis allows these special cause(s) to be identified and addressed as they arise.
Process mapping
The outcome of a successful process mapping exercise was to have developed a detailed description of the process under scrutiny.16 The final description must include every ‘work round’, variation and shortcut used in practice and, consequently, all members of the team should contribute to the mapping exercise. Process maps frequently demonstrate that in practice there is a lack of a standard approach to an activity.
Value stream mapping
This involves the examination of the process, often step by step, to identify low value, out of sequence and wasteful activities.17 It is also where unnecessary variability in a process is considered. The desired outcome of value stream maps was to describe, as a list, the potential things that might be changed, omitted or rearranged to achieve a more efficient process and, hence, a desired ‘future process’.
Failure Mode Effect Analysis (FMEA)
This is a risk assessment tool that can be used to quantify the relative risks associated with steps in a process.18 It requires a previous knowledge of how the system can fail (failure modes). These failure modes are then scored on the basis of how frequently the failure occurs, its severity, and how detectable the potential failure is before it occurs. The scale used is usually 1 to 5. The failure scores are multiplied together, and the resulting number referred to as the Risk Prioritisation Number (RPN). High RPNs indicate that a step in a process is highly unreliable and would require specific attention to reduce the associated risk or unreliability.
Matching capacity to optimise work flow (theory of constraints)
This involves the identification of constraints to work flow due to inappropriate allocation of staffing resource.19 The problem caused by ‘bottlenecks’ were often solved by simple reallocation of staff to match capacity and demand.
What to change and how
The application of the above techniques results in a list of potential changes that, if successfully implemented, might be expected to result in a more reliable, efficient and standardised process. Determining what is a ‘good idea’ is not straightforward. The implementation of a ‘good idea’ may result in unforeseen undesirable outcomes, and so each proposed change is tested to determine whether it will work in practice.
National medication safety alerts have highlighted the areas of practice that should be reviewed and they provide some recommendations. However, full process review and application of the techniques discussed have proven vital in ensuring objectively measurable improvements in areas, such as avoiding missed doses,20 are achieved.
Testing and the IHI model for change
Prior to its implementation, each proposed change requires testing. This was formally conducted using the IHI model of change.1 This is a modified Plan Do See Act (PDSA) cycle which forces the tester to formally clarify their outcome expectations prior to each test. This is achieved by asking three questions: ‘what are we trying to accomplish?’, ‘how will we know when a change is an improvement?’ and ‘what changes can we make that will result in improvement?’ This testing model was used to assess and record the outcome of each test. Unsuccessful or partially successful tests were modified and repeated. Eventually successful tests of change were implemented and became part of the new process.
Discussion
A positive attitude towards change now exists within the department. The application of the improvement methodologies above has led to a different way of viewing medicines management practices. Concepts of ‘waste’ and ‘value’ provide new ways for rationalising change. Once a system has been redesigned, its performance is almost automatically monitored. Pharmacy staff members are aware of the importance of operating standard systems and processes, and demonstrate the required discipline to achieve this.
While the earlier changes were implemented over a relatively long time period, later changes that employed formal methodology were implemented much more rapidly. The staff benchmarking data suggests that while the quality of service can be maintained, it is not necessarily dependent on the numbers of staff employed but in the way in which they work. This represents a potential saving to health communities.
The change management tools and techniques referred to above were relatively simple to use. Equally, once appropriate measures were chosen to monitor the effects of change, continued measurement and the interpretation of changes in these measurements was straightforward. Despite this, many improvement projects fail to achieve or maintain satisfactory outcomes. There are common reasons for project failure, and some of these are outlined below.
Improvement projects are more likely to succeed under the sponsorship of an engaged and influential champion. Such champions need not play an operational role in the project. However, it is useful if they have an understanding of improvement methodology. Their role is to provide support and, where necessary, promote the project, and to deal with organisational obstacles that would otherwise block progress.
Some projects fail because they choose inappropriate or overly complex metrics. Inappropriate metrics do not reflect change and, therefore, choosing the correct ones must be given careful consideration. Occasionally, a metric is chosen that is difficult or time consuming to collect. The team may therefore collect the data sporadically as it is seen as being an onerous task. The purpose of collating and maintaining robust and reliable data is to be able to demonstrate objectively that change has resulted in improvement (or not). It provides the measurable evidence for performance. It is therefore essential to ensure that data capture is as simple and relevant as possible, and that the responsible staff members demonstrate the discipline to maintain the required dataset.
Sometimes projects falter as a consequence of inadequate testing before implementation. Testing can go wrong in a variety of ways and it is important to accept that testing all proposed changes is essential before any attempt is made to implement them. Ideally, all initial tests should be carried out on a small scale and in a controlled environment to assess the concept of the proposed idea. Perhaps the most important factor to consider is the receptiveness of the test environment and how this may be optimised to increase the chances of a successful outcome. Key to this is the recruitment of enthusiastic participants into testing processes. This is referred to as ‘stacking the deck in one’s favour’, the rationale being that if a test of change could not be made to work in favourable conditions, it was not a suitable solution.
Project failure is also associated with too rapid implementation. Once the small-scale testing of change had proven successful, it is appropriate to implement it in a stepwise fashion. This approach is supported by the IHI1 and has proven more successful in achieving sustainability, when compared to a ‘big bang’ approach to implementation. Once the change had been implemented successful in one area it should be spread rapidly to other areas of the organisation, with the knowledge that it has been robustly tested. However, the possibility of further testing and minor adaptions of the process during the rollout stage should not be ruled out.
It is recognised that reliable processes are usually simple and relatively standardised. This is a fundamental characteristic of lean systems. High levels of variability are usually a characteristic of historical practices that have evolved over time. During the process and value stream mapping exercises, it was noted that some participants in these processes became uncomfortable at times as it required them to accept that the way in which they had been working may not be the most efficient, safe or effective. Some participants perceived this as being personally critical and resisted further involvement. To help avoid this, it was useful to fully explain the purpose of the activity and to anticipate and acknowledge this problem. An understanding of what constituted waste within a process was useful as was the realisation that that historic activity could be removed from work practices without detrimental effect.
With the arrival of Medicines Optimisation21 the question arises as to how small departments will address the new emphasis on patient experience, evidence-base intervention and safety for the individual patient. Managers of all hospital pharmacy departments are likely to be concerned about how these new challenges will be successfully addressed. Organisations might consider formally using the above change management techniques to improve efficiency and patient experience while at the same time being able to generate significant savings through the ability to redeploy staff.
Conclusion
This article shows that the use of common change management techniques recommended by the UK QIPP programme are applicable to the hospital pharmacy, and can result in measurable improvement in service delivery. It also indicates that the small size of a department is not necessarily a handicap to achieving better quality and efficiency in dispensary and chemotherapy production units, or in measurably addressing local and national patient safety concerns without the need for additional staffing resources.
What this paper adds
-
What is already known on this subject
-
Improvement methodologies adopted from the manufacturing sector for use in healthcare have been available to NHS professionals for a number of years.
-
With the introduction of the NHS Quality, Innovation, Productivity and Prevention (QIPP) programme in 2010, it is vital that pharmacy departments objectively demonstrate that they can embrace the available tools to bring about necessary change.
-
What this study adds
-
Evidence that common change management techniques and improvement methodologies are applicable to the hospital pharmacy, and can result in measurable improvement in service delivery.
-
Departmental size is not necessarily a handicap to delivering quality, safety and efficiency targets.
Footnotes
-
Contributors Article idea: JB; Literature search: DC; Authors: JB, DC, MA; Guarantor: JB.
-
Competing interests None.
-
Provenance and peer review Not commissioned; externally peer reviewed.