Faculty of Economics and Business SOM Graduate School EB
Nij Smellinghe
Technical Barriers to Implementation of Operational Performance Management Systems in Hospitals: A Case Study
Research Master, profile Operations Management Master of Business Administration, profile Business&ICT
08-December-2011 FINAL THESIS KONSTANTIN IGNATOV student number: 1747967 e-mail: k.k.ignatov@student.rug.nl Supervisors Dr. Tudor D. Bodea Dr. Taco van der Vaart Prof. Dr. Albert Boonstra
Acknowledgements
Abstract
Table of Contents
Introduction ... 1 Literature Review... 2 Method ... 4 Results ... 6 Hospital PMS Requirements ... 6 Objectives of the PMS ... 6 Terminology ... 7Hospital’s Care Process ... 8
Required Performance Measures ... 10
Technical Barriers to Implementation of PMS ... 11
Discussion ... 14
References ... 15
Appendix ... 18
Interviews Structure ... 18
Barriers to lean health care (de Souza and Pidd, 2011) ... 20
Overall Care Process ... 21
Introduction
Nowadays, hospitals are turning towards optimization of their internal operations through implementation of lean, six sigma and other methods in order to improve their efficiency and effectiveness, pressed by governmental budget cuts, increased costs and greater requirements for the quality of care (Rais and Viana, 2011, Jack and Powers, 2009, Li and Benton, 1996). To achieve their goals the hospitals re-design their facilities, business processes and information systems, whereby they strive to implement performance management systems (PMS) that can enable better decision-making and performance benchmarking (Rais and Viana, 2011) and can provide accurate information on markets, customers, competitive position, operational performance, etc. (Bititci et al., 2002). The term PMS is used in this study to depict a performance management system for hospitals which focuses on the operational performance. The system’s aim is to aid the operations managers to improve the resource utilization, throughput and waiting times without harming the clinical outcomes and without expanding the capacity by adding costly resources.
Performance management systems are complex systems involving design of performance measures at multiple organizational levels and implementation of information and communication technologies (ICT) (Nudurupati et al., 2011). Performance measurement involves three general stages – design, implementation and use of performance indicators (Nudurupati et al., 2011). The management of information systems plays a crucial role throughout the design, implementation and use of performance indicators (Nudurupati et al., 2011). De Leeuw and Van den Berg (2011) conclude that further research on the reasons for failures of PMS is needed in order to enhance the performance management practices and help the organizations avoid the danger of wasting time and financial resources on developing the wrong PMS. De Souza and Pidd (2011) explore the barriers to implementing lean health care through three case studies in the UK’s National Health System and compare them with the barriers found in manufacturing. They summarize the barriers in eight categories (shown in Table 7 in the appendix) and conclude that the adoption of lean health care is delayed by the barriers to implementation of PMS. In their paper, Clauser, Wagner, Aiello Bowles, Tuzzio and Greene (2011) examine the role of information technologies in improving cancer care. They conclude that the lack of standardized ICT-based “measurement and data collection that integrate seamlessly into practice” and “lack of guidelines on incorporating and using these measures in clinical workflow” are hindering the implementation and use of PMS (Clauser et al., 2011). Therefore, further research is needed to identify the potential sources/causes of the technology-related barriers to implementation of PMS and to develop and test solutions that can lift the technical barriers and enable more effective performance management in the health care.
systems are not well integrated? What can be done to improve the current situation and to enable a better performance management in the health care industry? These questions are discussed in the current study in order to provide with a roadmap to lifting the technical barriers to implementation of PMS.
This research paper presents a case study conducted within a Dutch hospital. The management of this hospital aims to implement a PMS in order to shorten the access times to medical specialists and the total lead times for patients, eliminate waste in the hospital’s processes, improve the resource utilization and respectively achieve higher quality of care at lower costs. Marley, Collier and Meyer Goldstein (2004) distinguish between clinical and process quality where the former “emphasizes ‘what’ the patient receives” and the latter “emphasizes ‘how’ health care services are delivered to patients”. The quality of care, in the context of this study, refers to the process quality (Marley et al., 2004). It is defined in terms of the lead times incurred by the patients and it does not include clinical measures such as mortality, infections after surgeries, etc.
The next chapter provides a review of the literature on performance management. Then the method used to conduct the case study is explained. Further, the paper shows the results from the analysis. Finally, the paper discusses the actions that can be taken by the hospital’s management in order to lift the technical barriers and enable the implementation of the required PMS.
Literature Review
This section begins with an introduction to the definitions of efficiency and effectiveness that are used throughout the study. The next sub-section discusses the literature on performance measurement and the issues associated with the definition and selection of valid and reliable performance measures. The literature on performance measurement is used to prepare the structures of the interviews carried out to collect the study data (see the next section Method and the Interviews Structure in the Appendix). The second sub-section was also used as a guideline during the first phase of the study, whereby the requirements for the PMS are investigated and the performance measures are defined. The third sub-section discusses the barriers to PMS found in previous studies.
Efficiency and effectiveness
Performance “is a function of both effectiveness [...] and efficiency” (Mentzer and Konrad, 1991). Effectiveness is defined as the “degree to which a goal is achieved” (Mentzer and Konrad, 1991). Efficiency refers to the performance of the business processes transforming inputs into outputs (Førsund and Hjalmarsson, 1974). Farrel (1957) discusses two types of efficiency – cost and technical efficiency. Technical efficiency measures “the success in producing maximum output from a given set of inputs” and cost efficiency measures the “success in choosing an optimal set of inputs” (Farrell, 1957). The term “efficiency” in this study is used in the sense of the “technical efficiency” defined by Farrell (1957). According to Førsund and Hjalmarsson (1974), the concept of efficiency is relative to the standards used as a basis for comparison and the selection of efficiency measures depends on the measuring purpose, thus, particular organizational objectives have to be defined when characterizing the firm’s efficiency.
Performance measures
accountability (e.g. Freeman, 2002, Li and Benton, 1996, Martinez et al., 2010). External summative performance measures require high data precision in order to allow for reliable comparisons of the performance of different organizations (Freeman, 2002). Internal formative use of performance measures supports continuous quality improvement and allows the management to gain better understanding of the actual operations in the organizations (Freeman, 2002). This research focuses on the internal use of the performance measures.
The involvement of the hospital’s management is an important element of the context in which a PMS is developed and deployed (Spath, 2007). An effective and efficient measurement system has to be flexible enough to accommodate the current and future requirements and this cannot be done without the active participation of the senior leaders of the organization (Spath, 2007). The performance measures should flow from the strategic objectives of the stakeholders (Spath, 2007). According to Freeman (2002), common understanding of the intended use of the PMS by all stakeholders is very important. In addition, the performance measures need to be selected so that the stakeholders have control over them as they are only meaningful when they represent constructs or process outcomes that can be influenced by the stakeholders (Freeman, 2002, Mc Kenzie and Shilling, 1998).
Martinez (2010) and Freeman (2002), among others, discuss the role of a performance review strategy in the context of using performance measures for internal continuous improvement. According to Martinez (2010), the performance indicators’ role to measure performance against explicit strategic objectives and to give feedback about the progress towards these objectives is fulfilled during the process of performance management review. Performance measures require constant re-evaluation to ensure their organizational relevance (Freeman, 2002), which makes it necessary to define a performance review strategy. Such strategy should address the questions of how often the performance indicators will be revised and how often the actual performance of the business will be evaluated based on the performance indicators.
Barriers to PMS
data element definitions, incomplete or missing data and variations of the coding practices between and within organizations. Bititci et al. (2002) also show that cumbersome and time-consuming, non-automated data collection is an important barrier to implementation of PMS. The technical barriers to PMS identified from the literature are summarized in Table 1. Despite the number of occurrences of the above technical barriers to implementation of PMS in the literature, it is still not clear why these barriers exist.
Table 1: Barriers to PMS from literature
Barrier Cause Source
Staff involvement and resistance Not reported Freeman (2002)
Lack of flexibility of the data collection and reporting systems; Non-automated data collection
Not reported Kennerley and Neely (2002); Bititci et al. (2002); De Souza and Pidd (2011)
Inflexible ERP systems Not reported Kennerley and Neely (2002)
Inappropriate “off the shelf” systems Not reported Kennerley and Neely (2002) Absence of “routinely available, clinically
detailed data”; Lack of data on the waiting times in waiting rooms
Not reported McGlynn (1997); Schneider et al. (1999)
Lack of systems linkage and data sharing Not reported Schneider et al. (1999) Lack of common definitions such as the
content of electronic medical records Not reported Schneider et al. (1999)
Data quality Data definitions, missing
data and variations of the coding practices
Schneider et al. (1999), Loeb (2004)
Security Not reported Schneider et al. (1999)
Method
A case study at a Dutch hospital was conducted for the period of six months in order to qualitatively investigate the causes of the technical barriers to implementation of PMS in hospitals. The hospital is of average size, having 281 beds, providing care for 28 medical specialties and realizing a throughput of about 111000 new patients per year. It uses a number of information systems for medical records, scheduling of patients, radiology and operating rooms, managing inventories, accounting, etc.
A qualitative case study approach was chosen for two reasons. First, as shown in the previous section, not much is known about the causes of the technical barriers to implementation of PMS and the case study approach is well suited for qualitative investigation of the relationship between the business context (technologies, practices, people, culture, etc.) and the barriers to PMS (Eisenhardt, 1989, Gibbert et al., 2008). Second, case studies represent “a methodology that is ideally suited for creating managerially relevant knowledge” (Gibbert et al., 2008), which is required for allowing the health care managers to pro-actively approach the problems related to PMS planning, design and implementation.
2002). To identify the barriers that are relevant for the hospital it is required to find out which performance measures can and which cannot be implemented (step 2). This, in turn, requires that the PMS measures needed by the hospital to achieve its operational goals are clearly defined (step 1). Thus, the PMS requirements were developed, which involved identifying the objectives of the hospitals’ managers and medical personnel, outlining the general care process used by the hospital and the sub-processes that needed to be measured, as well as how they should be measured.
Step 1: Interviews and development of the PMS requirements
Thirteen in-depth interviews (see Table 6 in the appendix for the positions of the interviewees) with the project stakeholders, the hospital’s internal documentation, information systems and direct observations were used in order to outline the general care process used by the medical specialists at the hospital and to identify the objectives of the management and medical personnel as well as the measures that can be used to benchmark the current state of the hospital against the objectives. The combination of the information sources listed above ensures the construct validity of the study (Eisenhardt, 1989, Gibbert et al., 2008). The interviews were recorded and then transcribed. Based on these data an overview of a complete PMS was drawn including all required performance measures and the ways they should be calculated and used. The interviews focused on clarifying the following elements, as identified from the literature (see section Interviews Structure in the appendix):
1. PMS Requirements
a. Managers’ and medical specialists’ strategic objectives and related measurable goals
b. PMS-related terminology
c. Perceived operational issues at the hospital, their causes and measurable effects
d. Controllable constructs in the hospital’s system and environment (scheduling, power of medical staff, etc)
e. Overall care process of the hospital 2. Barriers to implementation of the PMS
Step 2: Find out which measures can be implemented and which cannot
The next step of the research was to discuss the system requirements for the PMS with the ICT department at the hospital in order to check which of those requirements can be fulfilled in terms of data availability and current systems’ capabilities. This step resulted in a list of implementable measures for which data were available and a list of measures that required additional data to be collected before they can be implemented in the PMS.
Steps 3 and 4: Implementation and testing
The implementable measures were developed by using SQL (Structured Query Language) and integrated with the available reporting system at the hospital. Then, the resulting figures were analyzed in order to validate the consistency of the measures across all medical specialties and specialists and to check for existing outliers.
The last two steps involved discussing with the hospital’s managers and ICT staff the reasons for being unable to implement the PMS measures and what can be done to remedy the situation.
1. Carry on interviews and review the literature to compile the
requirements
PMS Requirements
2. Discuss requirements; Check what can be implemented
Requirements that cannot be implemented Requirements that can be
implemented
3. Write SQL scripts to obtain the raw data, calculate the required measures, integrate the results into
the current reporting systems
4. Test the measures, analyze the measures’ validity
Issues with validity of performance measures
5. Discuss with the hospital’s managers and ICT staff why the measures cannot
be implemented
6. Discuss with the hospital’s managers and ICT staff what can be done to lift
the barriers and implement the measures
Approaches to lifting the barriers
Legend
Process Step Document Output
Barriers and their causes to implementing the required measures
Figure 1: Case study protocol
Results
Hospital PMS Requirements
Objectives of the PMS
Improve the Market Position of
the Hospital
Operational Objectives
Utilization Throughput Lead Times
Costs Process Quality
Variability maximize
maximize minimize minimize minimize
minimize maximize
Quality of Scheduling
maximize maximize
Figure 2: Operational strategic objectives of the hospital Terminology
De Souza and Pidd (2011) suggest that the PMS-related terminology has to be clearly defined in order to ensure that there is a common understanding of the performance measures. Most of the interviewees found it hard to explain the differences between lead, access and waiting times. Some of them used all the terms interchangeably whereas others used the terms access and waiting times synonymously and the term lead times to describe the total time of care per patient. The interviewees had a common understanding of the terms throughput and utilization (Hopp and Spearman, 2008).
To achieve common understanding, the following definitions (also depicted in Figure 3) were obtained from the literature, diagrams depicting the patient flow were sketched and the beginning and ending points of the lead, access, process, cycle and waiting times were marked on the diagrams. All PMS stakeholders agreed upon the terminology and the meaning of the performance measures.
Lead Times – the total time of providing care to patients including the access and
cycle times per care problem. Hopp and Spearman (2008) distinguish between manufacturing and customer lead times. The manufacturing lead time “is the time allowed on a particular routing” whereas the customer lead time is the time “allowed to fill a customer order from start to finish” (Hopp and Spearman, 2008). The definition of lead times in this study refers to the customer lead times and can also be seen as the line cycle time in manufacturing – “the average cycle time in a line is equal to the sum of the cycle times at the individual stations less any time that overlaps two or more stations” (Hopp and Spearman, 2008)
Access Times – the time patients have to wait before they can go to the hospital
hospital it became clear that patients may experience long access times for some diagnostic procedures (e.g. MRI) and for some surgical procedures and that these access times are also very important determinants of the perceived quality of care. This made it necessary to include the treatment and diagnostics, in addition to the first consultations, in the definition of the access times.
Cycle Times – the sum of the waiting and process time per patient for each task
(e.g. waiting time for X-Ray + time for setup and taking an X-Ray photo + patient preparation time). The same definition is also used in manufacturing (e.g. Hopp and Spearman, 2008)
Waiting Times – the time patients spend in the waiting rooms at the hospital,
waiting for consultation, diagnostics or treatment. These times are measured in terms of minutes or hours. In manufacturing, the waiting times are also called queue time and refer to the time manufacturing jobs spend waiting for a station for processing (Hopp and Spearman, 2008)
Process Times – the time it takes to complete a task with the patient, e.g. surgery,
consultation, etc. The process time in manufacturing is the time jobs are being worked on at a station (Hopp and Spearman, 2008). In health care, it is important to distinguish between two types of setup times – patient and resource setup. The patient setup time involves preparing the patients for the medical procedures, thus, it is included in the process times, whereas the resource setup time involves preparation of the medical team and apparatus. The resource setup time, from the patient’s point of view, is actually waiting time, thus, it is included in the above definition of the waiting times
Lead Times Cycle Times (consultation, diagnosis, treatment) Access Times (consultation, diagnosis, treatment) + + Process Times Waiting Times + +
Figure 3: Hierarchy of time-based performance measures Hospital’s Care Process
adjusted to the overall care process for the given hospitals or care facilities. Nevertheless, this does not conflict with the objective of the study to examine the barriers to implementation of an operational PMS for internal continuous improvement. It has to be pointed out, though, that the overall care process and PMS must be well aligned.
Access times, hereby, are all waiting times incurred by the patients outside the hospital and are not included in the cycle times. There are access times for consultations, diagnostics and treatment, during which the patients wait at home.
The consultations refer to the moments when a patient visits a doctor at the outpatient clinic. The medical specialists distinguish between “first consultation”, “consultation” and “follow-up consultation”. The “first consultation” is used to check the patient’s status and symptoms and to plan the diagnostics process. The patients who have their first consultations are referred to as “new patients” because they enter the hospital’s system through the first consultation. The same patient can have more than one care problem and can be referred to the hospital separately for each care problem. Therefore, a patient can be a “new patient” more than once. The second, third, and so on consultations, until a diagnosis is established, are typically shorter than the first consultation and are used to review the diagnostic tests and to decide if more diagnostics is required or if a treatment can be prescribed. The “follow-up consultation” is done after some recovery time (after treatment) in order to check if the patient is recovering as expected.
The diagnostics is the process during which the patient can go to either radiology (X-Ray, MRI, etc.), laboratory (for blood testing, etc.) or to the functional ward (ECG, duplex, audio-metrics, etc.). Both the diagnostics access times and the waiting times in the waiting rooms can be sometimes very long, according to the interviewees. The access times can exceed several months for some medical specialties and the waiting times can exceed several hours.
hard for most of the interviewees to clearly distinguish between diagnostics and treatment. The end point of the treatment process coincides with the starting point of the recovery process. For the inpatients the treatment process ends when the patient is discharged from the hospital. For the outpatients the treatment ends when the patient is sent back home as well. Finally, the discharge process ends the treatment of inpatients and takes place when the doctors go to the wards in the hospital and decide which patients can be discharged (this is called “ward round”). After this moment the discharge has to be organized, e.g. transportation and tertiary care are arranged for. In some cases the access times for tertiary care can be very long, which significantly delays the discharge from the hospital causing loss of inpatient beds capacity. The times when the doctors do the ward rounds also have serious impact on the discharge process, as reported by the interviewees, because there may be no personnel to do the actual discharge at the time the ward round has been completed.
Required Performance Measures
The required performance measures were derived through analysis of the hospital’s objectives and overall care process and are summarized in Table 2. Five groups of measures were identified. The time-based measures include the lead times, access times, waiting times, process times and cycle times.
The throughput can have two general meanings in health care. First, there is the hospital throughput referring to the number of new patients that go through the care process per period of time. Second, there is the individual throughput of medical specialists and machine resources, which refers to the number of patients (regardless of whether they are new or not) that meet the specialists or use the medical apparatus per period of time. Additionally, the throughput of the medical specialists can be decomposed into the throughput of each type of activity that the given specialist carries out (e.g. throughput per specialists for new patients or for follow-up consultations).
The utilization refers to the utilization of machine resources and medical personnel. The variability is represented by the coefficients of variation of the measures (Hopp and Spearman, 2008). There are natural and artificial variability (Litvak et al., 2005). High variability implies less predictability and more allocated buffering resources (Hopp and Spearman, 2008, Litvak, 2005, Litvak et al., 2005). It is essential for the health care managers to measure the coefficients of variation in order to identify, analyze and eliminate the sources of artificial variability as well as to manage the natural variability.
Table 2: Required performance measures
Operational Performance Indicators
1 Time Based Indicators
Lead Times (LT) Access Times (AT) Cycle Times (CT) Waiting Times (WT) Process Times (PT) 2 Throughput (TH) 3 Utilization (U) 4 Variability (V) 5 Quality of Scheduling
Scheduled vs. Actual Activities (SAA)
Number of Canceled Consultations (CC)
Number of Canceled Surgeries (CS)
Number of Rescheduled Surgeries (RS)
The performance measures can be applied to different elements of the care process. The access times, for example, can be calculated for the first consultations, diagnostics, treatment and tertiary care. The lead times depend on the individual routing for each patient. Thus, some of the measures are resource-based whereas others are patient-flow-based. Table 3 presents the resources and their associated measures. In addition, it shows that the lead times, hospital throughput and their associated variability can be measured per patient flow. In total 60 measures are required in order to cover all operational aspects of the general care process of this hospital. Each of these measures is represented by two variables – a mean and a coefficient of variation, whereby the coefficient of variation represents the variability associated with the measure. In addition, the measures can be applied to individuals or groups of resources or medical specialists.
Table 3: Measures per resource and patient flow (see Table 2 for the abbreviations)
Resource Measure
First Consultation AT, WT, CT, PT, TH, U, SAA, CC Consultation WT, CT, PT, TH, U, SAA, CC Diagnostics
Laboratory AT, WT, CT, PT, TH, U, SAA, CC Radiology AT, WT, CT, PT, TH, U, SAA, CC Functional AT, WT, CT, PT, TH, U, SAA, CC
Treatment
Conservative AT, WT, CT, PT, TH, U, SAA, CC Surgical AT, WT, CT, PT, TH, U, SAA, CS, RS Discharge WT, AT for tertiary care
Patient Flow LT, TH
Technical Barriers to Implementation of PMS
“Elektronisch Zorg Informatie Systeem”). This system contains the electronic patient records (EPR), agendas of the medical specialists, planning of the operating rooms, etc. Another system is used for scheduling the radiology department. A third system is used for personnel management. There are also systems and sub-systems for accounting, shared reporting, database management, facility management and others.
encompass constructs that can be influenced by the stakeholders (Freeman, 2002, Mc Kenzie and Shilling, 1998) . Thus, the lack of data about the reasons for cancellations and waiting times prohibits the hospital’s management to properly analyze the hospital’s operations and to take actions that minimize the quality of scheduling performance measures.
Table 4: Technical barriers to PMS implementation
Barrier Attributed to Example
Lack of data Limitations of ICT used by the care provider no actual times, information about cancelations, rescheduling, diversions, etc.
Validity of measures Lack of policies for IS usage over-utilization due to agenda usage, non-standard codes used for all agendas
Lack of system
integration ICT development over long period of time
radiology has a separate agenda from all policlinic staff and resources
Lack of support from the ICT department
Management has not recognized the ICT department as a stakeholder in the PMS project
Lack of software tools required for the design, development and implementation of the PMS
Out of the 60 measures only 8 were implemented. They include the access times, throughput and utilization for first consultation, the throughput and utilization for consultations and the access times, throughput and utilization of the functional diagnostics. These measures were implemented because there were sufficient data in the EZIS. Still, there is one significant issue related to the utilization – it is not clear how to define the capacity and how to define the job done. Ideally, the capacity should equal the total planned minutes for a consultation type on a given date but the medical specialists are allowed to block parts (or entire periods) of their agendas so that no patients can be booked at the blocked slots. In this way the medical specialists are effectively reducing the capacity. The problem is that some specialists first block their agendas and then still book patients over the blocked slots without removing the blockages, which leads to very low capacity (due to blockings) and to very high utilization measures. In such cases the measure of utilization becomes invalid (Figure 5 in the appendix depicts such invalid measures, or outliers, due to inappropriate use of the agenda). With regard to the job done – two measures are possible given the available data in the hospital’s systems. First, one can use the booked minutes which do not necessarily represent the actual time a medical specialist has spent with a patient. Second, the worked minutes can be used but currently these data are not correctly recorded in the system and are only collected for the rheumatology specialty. Therefore, the actual minutes worked cannot be used at the moment but the PMS takes them into consideration to enable the future calculation of the utilization (based on actual minutes worked) when these data become available. Due to the above issues the PMS calculates four measures of the utilization of the medical specialists in the policlinic (they are summarized in Table 5):
Ufb: utilization based on full capacity (FC) and booked minutes (B)
Ufw: utilization based on full capacity and worked minutes (W)
Urb: utilization based on reduced capacity (RC) and booked minutes
Table 5: Measures of utilization
Booked Minutes Worked Minutes Full Capacity Ufb = B/FC Ufw = W/FC
Reduced Capacity Urb = B/RC Urw = W/RC
The lack of policies leads to differences in the usage of the agendas and obstructs the proper calculation of the utilization. This poses the problem of selecting the utilization measure with the greatest validity instead of using an actually valid standard measure.
Another technical barrier to implementation of the required PMS that was identified was the lack of system integration. For instance, the radiology ward has its own planning system that has a separate database and user interface. In order to implement the required measures related to the radiology it is necessary to develop a second PMS sub-system that pulls the data from the radiology database (which has different formats than the EZIS database), calculates the measures and merges them with the other measures. This results in two separate sub-systems that have to be maintained and adds to the complexity and costs of the PMS.
The last barrier to implementation of the PMS is related to the involvement and support of the ICT department. The ICT department is an important stakeholder in the PMS project and has to be recognized as such and involved in the project by clearly communicating the operational objectives and strategic plans to the ICT staff. During the course of the project, specific software tools for development were required but could not be provided by the ICT department for different reasons such as licensing and security. The objectives of the ICT staff are in practice conflicting with the PMS project’s objectives because the ICT department aims to reduce the number of information systems to be maintained and to increase the security of the current systems by limiting the access of the hospital’s personnel to them. On the other hand, the PMS project aims at implementing a new information system that accesses the data from the different patient-related databases. These conflicting objectives create barriers to the actual development of the scripts and reports and impose limitations to the PMS design. This problem is well recognized in the literature (e.g. Benson et al., 2004) but businesses still carry out planning, performance management and other activities in silos and do not connect their strategic decisions to the bottom-line operational goals and information systems.
Discussion
are not removed from the database and historical data about which consultation or surgery has been cancelled and re-scheduled are kept. Further, keeping record of the reasons for cancelations and re-schedulings may improve the validity of the performance measures. As to the waiting times, the actual times and durations of the consultations and surgeries need to be registered in the systems.
Better integration of the hospital’s information systems may enable the process of data collection and calculation of the performance measures, which is currently hindered by the multiple databases, technologies and applications used for scheduling and reporting.
The ICT department can be actively involved by the management in the PMS project in order to ensure the PMS is designed in the most effective way that enables easy maintenance and security of the system. The objectives of the medical, management and ICT personnel may be better reflected in the system by explicitly making the project stakeholders responsible for the design, development and implementation of the PMS.
Lifting the barriers “lack of data” and “low validity of measures” requires inter-organizational cooperation between the hospital and the EZIS provider in order to formalize the business processes and improve the capabilities of the EZIS system. This cooperation, though, can only be achieved after clarification of the responsibilities and costs associated with the development of the ICT systems, which may be hard to achieve due to lack of fit of the interests of both organizations and power misbalance. During the course of the project it was observed that the EZIS provider has gained considerable power over the decisions for what and when to be implemented in the EZIS because their system has become deeply intertwined within the hospital’s processes, which has increased the dependability of the hospital on the specific software. Therefore, it might be better for the hospital to find a way to gain the active involvement of the EZIS provider in the business process analysis and formalization and to work with the provider as a partner rather than as just a supplier.
Finally, this case study involves only one hospital, which is a limitation of the research. Despite this limitation, it is expected that other hospitals face the same problems for two reasons. First, the historical development of the hospitals’ physical and management structures (functional silos) and the power that the ICT providers have gained over the years are similar among different hospitals. Second, particularly in the Netherlands, a number of hospitals use the same EZIS. Moreover, before being able to conduct a proper empirical quantitative research on the causes/remedies of the barriers to PMS in health care it is necessary that a body of scientific literature is built on the topic. The current study adds to this body of literature in order to generate deeper understanding of the relationships between the elements of the systems’ context and the PMS implementation in health care. This, effectively, may result in improvements of the performance of hospitals and, therefore, in reduction of the health care associated costs.
References
BENSON, R. J., BUGNITZ, T. L. & WALTON, W. B. 2004. From Business Strategy to
IT Action.
BITITCI, U. S., NUDURUPATI, S. S., TURNER, T. J. & CREIGHTON, S. 2002. Web enabled performance measurement systems: Management implications.
CLAUSER, S. B., WAGNER, E. H., AIELLO BOWLES, E. J., TUZZIO, L. & GREENE, S. M. 2011. Improving Modern Cancer Care Through Information Technology. American Journal of Preventive Medicine, 40, 198-207.
DE LEEUW, S. & VAN DEN BERG, J. P. 2011. Improving operational performance by influencing shopfloor behavior via performance management practices. Journal of
Operations Management, 29, 224-235.
DE SOUZA, L. B. & PIDD, M. 2011. Exploring the barriers to lean health care implementation. Public Money & Management, 31, 59-66.
EISENHARDT, K. M. 1989. Building Theories from Case Study Research. The Academy
of Management Review, 14, 532-550.
ELKHUIZEN, S. G., DAS, S. F., BAKKER, P. J. M. & HONTELEZ, J. A. M. 2007. Using computer simulation to reduce access time for outpatient departments.
Quality and Safety in Health Care, 16, 382-386.
FARRELL, M. J. 1957. The Measurement of Productive Efficiency. Journal of the Royal
Statistical Society. Series A (General), 120, 253-290.
FOLAN, P. & BROWNE, J. 2005. A review of performance measurement: Towards performance management. Computers in Industry, 56, 663-680.
FØRSUND, F. R. & HJALMARSSON, L. 1974. On the Measurement of Productive Efficiency. The Swedish Journal of Economics, 76, 141-154.
FREEMAN, T. 2002. Using performance indicators to improve health care quality in the public sector: a review of the literature. Health Serv Manage Res, 15, 126-137. GIBBERT, M., RUIGROK, W. & WICKI, B. 2008. What passes as a rigorous case
study? Strategic Management Journal, 29, 1465-1474.
HOPP, W. & SPEARMAN, M. 2008. Factory Phisics, McGraw-Hill.
JACK, E. P. & POWERS, T. L. 2009. A review and synthesis of demand management, capacity management and performance in health-care services. International
Journal of Management Reviews, 11, 149-174.
KENNERLEY, M. & NEELY, A. 2002. A framework of the factors affecting the evolution of performance measurement systems. International Journal of
Operations & Production Management, 22, 1222-1245.
LI, L. X. & BENTON, W. C. 1996. Performance measurement criteria in health care organizations: Review and future research directions. European Journal of
Operational Research, 93, 449-468.
LITTLE, J. D. C. 1961. A Proof for the Queuing Formula: L= λ W. Operations Research, 9, 383-387.
LITVAK, E. 2005. Optimizing patient flow by managing its variability. Front Office to
Front Line: Essential Issues for Health Care Leaders. Oakbrook Terrace, IL: Joint Commission Resources, 91-111.
LITVAK, E., BUERHAUS, P. I., DAVIDOFF, F., LONG, M. C., MCMANUS, M. L. & BERWICK, D. M. 2005. Managing Unnecessary Variability in Patient Demand to Reduce Nursing Stress and Improve Patient Safety. Joint Commission Journal on
Quality and Patient Safety, 31, 330-338.
LOEB, J. M. 2004. The current state of performance measurement in health care.
MARLEY, K. A., COLLIER, D. A. & MEYER GOLDSTEIN, S. 2004. The Role of Clinical and Process Quality in Achieving Patient Satisfaction in Hospitals.
Decision Sciences, 35, 349-369.
MARTINEZ, V., PAVLOV, A. & BOURNE, M. 2010. Reviewing performance: an analysis of the structure and functions of performance management reviews.
Production Planning & Control, 21, 70 - 83.
MC KENZIE, F. C. & SHILLING, M. D. 1998. Avoiding Performance Measurement Traps. Compensation & Benefits Review, 30, 57-65.
MCGLYNN, E. A. 1997. Six challenges in measuring the quality of health care. Health
Affairs, 16, 7-21.
MENTZER, J. T. & KONRAD, B. P. 1991. An efficiency/effectiveness approach to logistics performance analysis. Journal of Business Logistics, 12, 33-62.
NUDURUPATI, S. S., BITITCI, U. S., KUMAR, V. & CHAN, F. T. S. 2011. State of the art literature review on performance measurement. Computers & Industrial
Engineering, 60, 279-290.
RAIS, A. & VIANA, A. 2011. Operations Research in Healthcare: a survey.
International Transactions in Operational Research, 18, 1-31.
SCHNEIDER, E. C., RIEHL, V., COURTE-WIENECKE, S., EDDY, D. M. & SENNETT, C. 1999. Enhancing Performance Measurement. JAMA: The Journal
of the American Medical Association, 282, 1184-1190.
SPATH, P. L. 2007. Taming the measurement monster. Frontiers of health services
management, 23, 3.
WHOLEY, J. S. & HATRY, H. P. 1992. The Case for Performance Monitoring. Public
Administration Review, 52, 604-610.
YIN, R. 2002. Case Study Research: Design and Methods, Third Edition, Applied Social
Appendix
Interviews Structure
Table 6: Interviewees Positions
Position
Management and administration
Director
Manager surgical care Manager non-surgical care Operations manager ICT system administrator Researcher/Expert
Medical
Head non-surgical outpatient clinic Head IC/CC
Cardiologist
Gynaecologist; medical advisor to board and chairman of medical staff Coordinator functional department (ECG, etc.)
Head A2; internal medicine Head short-stay and day care
Interview objectives:
1. Outline and formalize the long-term and short-term strategies
2. Identify and formalize the objectives of the interviewees with respect to their functions/positions in the hospital
3. List the operational/logistic problems according to the interviewees and with respect to their functions/positions in the hospital
4. Define the controllable measures/indicators of the operational problems 5. Draft the care sub-process pertaining to the interviewee’s ward or specialty
Interview structure
1. Background information to be given/explained to the interviewees (10min) a. Give an introduction of the project
b. Explain the difference between internal/external PIs and ensure the interviewees understand and agree upon the use of the PI system
c. Explain why it is important to link the strategic objectives to the performance indicators and to the goals
2. Questions (50min)
a. Are there formal long-term and short-term strategies that guide the management? If yes, what are the main strategic objectives?
b. What is your function/position/responsibility in the organization?
d. Is all of the information that your tasks require available whenever you need it? If not, what extra information do you need, what for, and how often/when? If yes, do you have any suggestions to improve the information’s quality and/or availability?
e. Elaborate on the additional information and the source of the problem. Is the currently available information accurate and reliable?
f. What problems do you face in performing your job/tasks? E.g. do you have the necessary space, resources, time and information whenever you need them?
g. Can you suggest some indicators of the problems?
Barriers to lean health care (de Souza and Pidd, 2011)
Table 7: Barriers to lean health care (de Souza and Pidd, 2011)
Barrier Evidence Incidence
Perception
Manufacturing myths and lack of understanding of lean principles among health care professionals is seen as a barrier
(H)—Unique to health care
Terminology
Introduction of new language is a common issue for implementing lean in any setting. In general, health care professionals responded well to the introduction of new vocabulary and it helped them to shift from old to new practices (M and H)—Common to manufacturing and health care Personal/professional skills of health care professionals
There are intrinsic differences in personal and professional skills between health care and manufacturing professionals and these differences are seen as a barrier. In the particular case of the NHS, it becomes clear that the fire-fighting mentality acts as a practical barrier in the introduction of lean
(H)—Unique to health care
Organizational momentum
The constant change of strategy for improvement (locally) and governmental policy (nationally) inhibits the continuity of potentially successful programmes
(M and H)—Common barrier but emphasised in health care due to its complexity
Professional and functional silos
The fragmentation of health care into silos (professional or functional) imposes a major barrier to the flow of patients, goods and information and consequently to the
implementation of lean techniques in hospitals
(M and H)—Common barrier but emphasised in health care due to its complexity
Hierarchy and management roles
Cultural issues based on the hierarchy of health care staff and the way management roles are allocated typically become a barrier for any improvement but this is especially important when lean is introduced
(H)—Unique to health care and also frequent in the public sector
Data collection and performance measurement
Lean implementation usually reveals problems in data collection and poor performance measures in most aspects of patient care. This often amplifies the need for cultural change in health care settings
(M and H)—Common barrier but emphasised in health care due to its complexity
Resistance to change/scepticism
Resistance to change is a significant problem in any improvement programme in any organization. It deserves special attention from those attempting to implement lean, since staff empowerment, which is a key issue in the lean theory, is needed for engaging health care professionals
