Chapter 2. Operations management: the search for value in healthcare organisation and performance
Sandra Leggat
Introduction
If corporations have glass ceilings, then hospitals have concrete floors. What happens above in the general management seems awfully disconnected – in activity if not quite in consequences – from what happens below in clinical operations.
(Mintzberg 2002:197)
In most organisations a set of activities creates value by transforming inputs into outputs. The outputs are products or services. Operations management, broadly defined as the planning, management and control of these activities (‘the operations’), is a fundamental component of management practice. Operations research, management science or industrial engineering apply mathematical or statistical models to operating issues to guide management practice and improve operations. In many industries operations management has been key to productivity and quality improvements.
Effective management of clinical care processes, which are the fundamental operations of a healthcare organisation, is essential for a well-functioning healthcare system. Yet it is only recently that the healthcare sector has recognised the relevance of operations management. Well-established operations management approaches, such as six sigma; lean thinking; root cause analysis; failure mode and effect analysis; queuing theory; modelling and simulation; and supply chain logistics, have become recent additions to the health services management repertoire (see Warburton, Chapter 9).
This chapter provides the context for taking up operations management in managing clinical processes in healthcare. ‘Traditional’ clinical and managerial processes have been accepted as the basis for healthcare service planning, delivery and evaluation. But, as outlined in Chapter 1, healthcare systems throughout the world are experiencing pressure to change – consumer influence, competition, changes in public policy, and advances in technology and clinical practice require review of traditional processes. Effective clinical process management requires understanding of those variables that have the power to significantly improve healthcare processes, and ultimately the performance of the healthcare system. This chapter explores the reasons why process management is difficult in healthcare and suggests future directions to improve clinical care processes.
Operations management – a sound management discipline
Operations management originated with the industrial revolution in the 1700s. Up until this time consumers received their products and services from individual craftspeople. While the individual form of craft production is still evident today, the industrial revolution facilitated the widespread production of consumer goods. This mass production was based on inventions ranging from the division of labour, Eli Whitney’s introduction of standardised parts in 1790, the principles of scientific management and the invention of the computer, which allowed complex analysis and modelling. These foundations of mass production enabled significant savings through economies of scale, largely achieved through standardising tasks to shorten production time and reduce variation and human error.
Unfortunately mass production also created problems in defining, measuring and ensuring quality. No longer was one craftsperson accountable for, and in control of, the product quality throughout the production process. Early approaches to quality control focused on finding defects and removing them at the end of the production process. As knowledge of production processes improved, scholars suggested that this inspection-based approach could be replaced with error prevention. In the early 1920s statistical process control was introduced to manufacturing processes to reduce errors. W Edward Deming used statistical process control and his knowledge of management to improve production in the US during World War II, and then introduced statistical control to Japanese manufacturers. Joseph Juran added a human dimension to quality control, suggesting the need for managers to be trained in quality methods. This requirement of training was not accepted in the US and Juran became known for his success in improving Japanese quality control after the war (see also Boaden & Harvey, Chapter 10).
The next wave of operations management innovation focused on lean manufacturing with just-in-time systems and processes. This meant that producers no longer had to store large inventories and products were ‘pulled’ through only those processes that added value, based on customer demand (Bowen & Youngdahl 1998). Efficiencies were achieved by grouping inputs with similar process requirements and by reducing set-up times. Importantly, lean structures shifted the responsibility for quality control from inspectors and quality departments to individual workers and teams (Bowen & Youngdahl 1998). These advances in operations management illustrate an ongoing cycle where a product is introduced and then continuously improved until a ‘final’ ultimate product is achieved. Once the preferred design has been identified, the focus of innovation moves to the production processes. In comparison with manufacturing, the health service sector has not found either the perfect design or the most appropriate processes.
Operations management – necessary, but insufficient in healthcare
Although operations management has been around since the industrial revolution, in healthcare we have been slow to capitalise on the analysis techniques and improvement methods of this discipline to manage clinical operations. There are some examples: In the late 1990s American hospitals advanced clinical process improvement and innovation (CPI) to decrease costs and improve quality (Savitz 2000) and by 1995 over 60% of US hospitals reported use of clinical process management (Walston et al 2000). But in comparison with other industries, where processes are designed to operate correctly 98–99% of the time, in healthcare studies have shown that clinical care processes may be defective 50% of the time (Resar 2006, Scott et al 2004, Runciman et al 2006). While healthcare has assumed some aspects of effective operations management, it is patchy and sub-optimal in comparison with other industries. Specialisation and standardisation are two principles that have transformed the operations of many industries but which have had mixed success in healthcare. Each is discussed briefly in the context of healthcare.
Specialisation in healthcare
There is an over-abundance of specialisation in workforce inputs in healthcare. Specialisation, or division of labour, defines the extent to which tasks in an organisation are subdivided into separate jobs. In healthcare, this is commonly based on professional expertise and clinical specialties. More than one hundred health professions can be identified, and within only one of these professions there are over 130 medical specialties. In Australia, while we continue to support this abundance of specialists, we have difficulty maintaining adequate numbers of practitioners within many individual specialties. The increasing specialisation of healthcare has required mechanisms to facilitate interdependence among the workers involved in a patient’s care. The boundaries of these professional groups and an individual’s hierarchical rank reduce information sharing (Edmondson 1996) making it difficult to manage care processes effectively. This has led to the hospital being referred to as the ‘key battleground for the various forces arrayed in the division of labour in healthcare’ (Dingwall et al 1988:228). While it is acknowledged that ‘quality comes from improving processes, which invariably cut across professional and functional boundaries’ (Buck 1998:752), the organisation of healthcare has not kept pace. The necessary inter-disciplinary and inter-departmental coordination are neither encouraged, nor rewarded.
In other industries specialisation has been effective in improving production processes, with a focus on ensuring specialisation of inputs other than the workforce, as well as specialisation of processes and outputs. While there is abundance of specialisation among healthcare providers, researchers have commented that unlike other industries, specialisation in other inputs (such as equipment, facilities, and even patients) has not been seen to be linked to improved clinical outcomes (Ramanujami & Rousseau 2006:823). Most recently ‘lean thinking’ approaches have been adapted from car manufacturers to healthcare settings and have resulted in an increasing focus on structuring service delivery to increase specialisation among patient inputs (Ben-Tovim et al 2007, Kelly et al 2007). The concept of the hospital as a ‘focused factory’, with highly specialised services consolidated in a relatively small number of sites, and with greater differentiation of patient types between emergency-driven hospitals and elective care (Leung 1999), is a good example of how specialisation of manufacturing production process can be adapted to the hospital setting.
Standardisation in healthcare
Standardisation has offered savings through efficiencies in many industries. Standardisation is broadly defined as the process of establishing agreed technical standards to achieve benefits for an organisation or industry. Standardisation can be applied to all parts of production – the inputs, the production processes and the outputs – and defines the specifications of the components in these parts, that is, the desired technical standards. In healthcare, like specialisation, standardisation is primarily directed to the competencies of health professionals (Glouberman & Mintzberg 2001b), with professional colleges and registration bodies structured to ensure standards of practice. While this aim has ensured standardisation of the largest proportion of the inputs, standardisation of other inputs (such as equipment and supplies) has not reached the levels achieved in other industries that have enabled efficient operations. Recently, standardisation of other inputs in healthcare has increased patient safety, as well as contributing to efficiency. For example, standardising medical supplies has enabled staff to rationalise their knowledge of the application of a wide variety of similar supplies, thereby assisting to improve error-free use. Standardising medical supplies also improves purchasing power.
Care pathways and clinical guidelines are examples of standardisation of the production processes (‘the work’) (see Claridge & Cook, Chapter 4). Recent initiatives, such as the UK National Health Service’s (NHS) Map of Medicine, employ process management techniques, almost by stealth, to introduce greater standardisation in the craft-process relationship between clinician and patient. However, doctors relish their individual independence and reject standarisation of care (Degeling et al 2001). Thus, while there is evidence that the standardisation offered through pathways and guidelines improves quality of care (Caminiti et al 2005, Vikoren et al 2006) there has been mixed success in implementation and consistent use, with recent studies identifying substantial barriers. Within healthcare, the difficulties in standardising work reflect the
… insoluble paradox between the need for consistent and evidence-based standards of care and the unique predicament, context, priorities, and choices of the individual patient.
(Plsek & Greenhalgh 2001:627)
This paradox distinguishes health service management from management in other industries, as health is characterised by:
▪ complex decision making at the patient care level that is negotiated between the patient and healthcare professional
▪ serious consequences of errors in decision making that may result in death or injury
▪ an uncertain external environment, with the combination of public and private financing and delivery schemes making it difficult to navigate
▪ goals of service delivery, which are often ambiguous and potentially conflicting (Leatt & Porter 2003).
While these factors might suggest that healthcare by definition will have high variation, studies consistently confirm that the variation caused by the organisation of the delivery system outweighs the variation caused by the random arrivals of patients with their ‘unique’ needs (Haraden & Resar 2004). The need to balance standards of care with patient individuality may also explain the fact that in comparison with other industries, there has been little emphasis on standardisation of outputs in healthcare (Glouberman & Mintzberg 2001b). The reduction in anaesthesia deaths in the US from rates of 25 to 50 deaths per million in the 1970s and 1980s (Ross & Tinker 1994) to current rates of less than 5 per million (Eichhorn 1989) and the development of diagnosis-related groups are two examples of the standardisation of outputs, although it is difficult to find others.
Operations management – the obstacles in healthcare
Three hypotheses are proposed to explain why operations management has not permeated the healthcare industry and are discussed further below:
1. the influence of craft production in a mass production environment
2. the difficulties in applying production management concepts to services
3. the emphasis on inputs in public administration.
The influence of craft production in a mass production environment
Healthcare has tended to be craft-based production – a trained healthcare professional provides his or her craft for individual patients, with little need for management. But as healthcare has moved from service delivery primarily in the community (that is, in the patient’s home or the doctor’s local surgery) to institutions such as hospitals, aspects of mass production have been introduced.
The organisation and operation of the hospital illustrates the difficulties transferring craft production to a setting that requires teamwork, coordination and integrated production. Hospitals display a fundamental inconsistency in that they are organised in formal managerial and clinical hierarchies that are based on scientific management principles, yet they try to maintain a commitment to professional autonomy for clinicians (Leggat & Dwyer 2005). The complex multifaceted production processes in hospitals are managed with a craft production mentality. This has resulted in an emphasis on managing (through the hierarchies) those aspects of operations that don’t interfere with the craft production relationship between clinician and patient. This was demonstrated by a review of management decisions in the NHS which found little managerial control over medicine (Harrison & Lim 2003), and by an Australian study that found that clinician managers focused on financial management, people management, organisational management and customer orientation (Braithwaite 2004). Clinical management was not identified as a primary pursuit of this group of managers. Instead, process, quality and data management – key components of clinical management – were only found in the secondary pursuits, where the clinician managers reported spending less time and effort (Braithwaite 2004). A more recent study of Australian public health managers found similar results (Braithwaite et al 2007).
Avoiding management of the clinician–patient relationship has resulted in independent, and largely inefficient, craft production. Instead of an effective interdisciplinary care delivery model, hospital organisation and hierarchy reinforces parallel care processes that only occasionally intersect. In addition, the healthcare professions have differing views on the evidence for effective practice, and the education, training and work practices of our health workers provide limited opportunities for the multidisciplinary evidence sharing and debate necessary to achieve consensus on clinical processes (Dopson et al 2002). When care is delivered through multiple clinical processes that are based on different clinical evidence and that only occasionally intersect, management of the processes to achieve efficient, consistent and high-quality care is difficult.
The healthcare craft production model depends on the staff involved to deliver error-free service. Quality control is largely focused on post-process audit. Other industries have realised that human beings cannot consistently maintain the required high performance levels and therefore create systems to reduce variation within processes. These systems anticipate and compensate for the likely errors that normal humans make (Resar 2006).
Pause for reflection
Why do healthcare managers in many countries struggle to find an appropriate operating structure that can influence service quality in a craft production environment? Why, as a result, is the healthcare industry characterised by a tension between craft production and mass production operation and quality control principles?
The difficulties in applying production management concepts to services
Services and products are different. Because operations management has evolved from production processes it has taken time to see how the concepts might translate to service provision, such as healthcare. While General Electric (GE) was able to quickly use a six sigma strategy to set reliability goals for manufactured products, it took much longer to apply the same concepts to the GE service-oriented businesses. Unlike manufacturing, service production and consumption take place simultaneously with high customer interaction, but the operations management concepts are the same. For example, while production control in other industries is concerned with the movement of materials, in healthcare the flows of the patients are critical to the process (Vissers & Beech 2005).
Evidence from the Institute for Healthcare Improvement suggests that improving patient flow can enhance clinical outcomes, address patient safety, increase patient and staff satisfaction and reduce operating costs (Haraden & Resar 2004). Yet application of process management to improve patient flow (from the perspective of both the patient and the providers) has been difficult, as no one has responsibility for the patient throughout their entire journey. As described by Glouberman & Mintzberg, within a hospital the medical staff tend to have discrete interventional relationships with patients, while the nursing and allied health staff play a more continuous part in the production process (Glouberman & Mintzberg 2001a). However, the nurses and allied health practitioners ‘are functionally subordinate to the physicians’ (Glouberman & Mintzberg 2001a
Buy Membership for Internal Medicine Category to continue reading. Learn more here
