Learning for the NHS on procurement and supply chain management: a rapid evidence assessment

Policy context

In the UK, the government has made reducing the public deficit its greatest priority, with significant implications for public sector services including health care. 1 Following a decade of growth in National Health Service (NHS) funding, this has slowed substantially from 2011–12. 2 This is set against a need to make efficiency savings of £20B by 2014–15 while improving the quality and delivery of NHS care. 3 NHS organisations are therefore under considerable pressure to contain cost while at the same time meet the growing demand for health care and ensure the quality of treatment and care. 3

There are various options by which the efficiency of the health system and of organisations operating within it may be enhanced. One way of thinking about this is to differentiate between types of inefficiencies that occur at different levels in the health-care production process, considering operational, allocative and administrative processes. 4^,5 Thus, operational inefficiencies occur because of duplication of services and inefficient processes, the use of expensive inputs or errors. Allocative inefficiencies result from misalignment of resources against best possible outcomes that could be achieved. Measures to strengthen allocative efficiency would involve rebalancing services across the health system, improving care co-ordination or strengthening preventative care. 5^,6 Administrative inefficiencies occur as a consequence of administrative spending which exceeds that necessary to achieve the overall goals of the organisation or system,4 also referred to as ‘back office’ functions. 4,7 Improving administrative efficiency could be achieved through, for example, (de)centralising administrative functions, simplifying administrative procedures and introducing uniform standards.

Ongoing activity in the NHS is seeking to address these different types of inefficiencies in different ways, with the potential for savings in operational and administrative functions. In particular, procurement and back office are seen as important areas to achieve efficiency gains. 2 One area that has come under scrutiny is NHS trusts’ non-pay expenditure, which, on average, accounts for around 30% of their total expenditure. 8 In 2011–12, this expenditure was estimated at £20.6B, of which over one-quarter was spent on drugs and pharmacy and just over one-fifth on clinical supplies and services (Table 1). 9

A 2011 review of NHS spending on medical and other consumables found wide variation in purchasing across acute trusts in England, with differences in processes and product ranges procured, alongside variation in prices paid for the same items. 10 The review highlighted the scope for efficiency savings in this area of spending. It estimated the potential of savings to be £500M, which equates to 10% of the annual spend on NHS consumables. Areas identified as offering potential for even greater savings include strengthening the strategic vision for purchasing and logistics.

Procurement in the NHS

Procurement of medical supplies and other consumables by NHS trusts in England can be realised in various ways. NHS trusts can, individually or in collaboration with others, directly contract with suppliers, draw on the national supplies organisation (NHS Supply Chain) or use one of the nine regional collaborative procurement hubs. 10 These hubs are regional purchasing organisations that were introduced during the 2000s in an attempt to achieve savings for partner trusts at regional level through aggregating their procurement efforts. 11 NHS Supply Chain was formed in 2006 from the NHS Logistics Authority and parts of the NHS Purchasing and Supply Agency, which had been established in 2000 as an executive agency of the Department of Health. 12 Operated by DHL Supply Chain Limited and managed by the NHS Business Services Authority on behalf of the Department of Health, it provides, at national level, procurement and logistics customer and supplier support. 12 Purchasing decisions are otherwise controlled by individual trusts via the local route of direct contracting between individual providers and suppliers.

More recently, the 2008 Procurement Capability Review of the Department of Health and the NHS,13 conducted by the Office of Government Commerce, identified weaknesses in approaches to procurement in the NHS, for instance lack of agreed strategy and operating model. In response, and as part of its wider strategy for the NHS, the Department of Health set out a new commercial operating model for the Department of Health and the NHS. 14 Among other things, this included the introduction of regional commercial support units (CSUs), which were intended to provide commercial support to both NHS commissioning and provider organisations, and were expected to work alongside NHS Supply Chain to ensure value for money for goods and services procured (Table 2). There was also an expectation for CSUs to merge with the corresponding regional collaborative procurement hub. The new model further involved the dissolution of the NHS Purchasing and Supply Agency, and its responsibility for procurement policy and pharmaceuticals procurement was transferred to the Department of Health. 17 At the same time, and in response to Innovation Nation (2008),18 which committed each government department to include an innovation procurement plan as part of its commercial operating model, the Department of Health published the National Innovation Procurement Plan (2009). 15 It highlighted the importance of innovation procurement in safeguarding quality, productivity and sustainability in the NHS. The plan sought to provide a coherent framework for innovation procurement by organising the adoption of technology-led innovation at the regional level, supported by an innovation fund to promote faster innovation and more universal diffusion of best practice. These overall developments took place against a wider reform programme of the NHS that sought to enhance patient choice and competition between providers while emphasising the need to secure quality, innovation and productivity. 19

Recognition of the importance of procurement as a means to drive up quality and value, and to stimulate innovation in the NHS more widely, was further emphasised in the 2011 Innovation Health and Wealth: Accelerating Adoption and Diffusion in the NHS report,16 which also noted how procurement would provide an important lever for economic growth. It announced the publication of a procurement strategy by the Department of Health in 2012, which would include a range of measures to help reduce waste and achieve efficiency gains in procurement as identified by the National Audit Office (2011), as described earlier. 10 This strategy was eventually published in August 2013, although referred to as a ‘procurement development programme’ rather than a strategy as such. 9 Highlighting how non-pay expenditure in NHS trusts had continued to increase over time, and at a rate higher than NHS activity and inflation during 2011–12, it set out a programme of work that seeks to stabilise non-pay spending for the period until the end of 2015–16. It proposed four core initiatives that aim to (i) deliver immediate efficiency and productivity gains; (ii) improve data, information and transparency; (iii) revisit the nature of clinical engagement in procurement; and (iv) create a national ‘enabling function’ to support leadership and build procurement capability throughout the system (see Table 2). We will return to the 2013 procurement development programme in subsequent sections of this report as it provides important context for the findings presented here.

Informing NHS learning for procurement

In its 2012 review of progress made in the NHS towards achieving efficiency savings, the National Audit Office highlighted the need for robust evidence to help the NHS make informed decisions about how to make such savings. 2 It pointed to the potential for lessons to be learned from activities and initiatives implemented elsewhere to enable the adoption of good practice.

In the field of NHS procurement, and supply chain management (SCM) more generally, there is potential to learn from other sectors, public and private, and from experiences in other countries, and to identify potential for cost containment and efficiency gains. A growing body of work has studied the applicability of SCM concepts developed in the private sector to public services to inform better use of public funds. 20 Much of the existing literature which draws on good practice in the private sector focuses on manufacturing, with the automotive sector one of the most studied industries. For example, Toyota’s ‘just in time’ supply management model, which aimed at improving return on investment and limiting inventory costs, has been explored in some depth for its transferability to other sectors. 21 The Toyota model relies on developing close links with a small number of suppliers, level production scheduling and continuous quality improvement,22 and this has also informed discourse and practice in health-care settings. 23–25

The public sector may also offer opportunities for learning. For instance, following a review of UK Ministry of Defence routine procurement items in 2007,26 the ministry introduced a number of measures intended to streamline processes and improve the cost-effectiveness of routine procurement by introducing measures such as e-procurement and reverse auctions, and changing some of the low-cost supply routes for routine items. 27 There is also potential for learning from experiences of procurement in health systems other than the English NHS. Countries that may provide useful insights into procurement and SCM in the health-care sector are Italy28 and New Zealand,29 owing to their recent reforms to strengthen efficiency in health-care procurement.

Although available evidence provides potential for models developed in other sectors to be adapted in health-care settings, there is a need to bring together the diverse literature on such approaches that may be relevant to the NHS context. Work that is available has highlighted that the NHS has a substantial potential to influence the supply chain in some of the products it purchases. 30 The 2011 review by the National Audit Office of NHS procurement of consumables argued that more efficient procurement has the potential to save costs without reducing the quality of patient care. 10 At the same time, lessons learned will have to be placed in the wider context of the quality improvement agenda31 and the need to create value in health care. 32 There is therefore a need to better understand the potential of new approaches used in other sectors to inform decision-making in the NHS, and the risks and benefits associated with these.

The work presented in this report seeks to contribute to this process by advancing our understanding of the evidence on procurement and SCM in sectors other than health care that can inform practice in the NHS. Principally drawing on a rapid evidence assessment (REA), we sought to

1. describe approaches to procurement and SCM in selected areas (including, but not limited to, manufacturing and automotive sectors, defence, information and communication technology and pharmaceutical industries)

2. identify best practices that may inform procurement and SCM in the NHS.

Search strategy

The literature on procurement and SCM stretches beyond peer-reviewed journals to trade publications and government reports. Given the highly theoretical nature of most of the academic publications, we included examples from smaller studies or empirical data from case studies in practice, regardless of whether this was academic or grey literature.

Our pilot testing of the search terms underwent several iterations, and the search was conducted in three stages, focusing on (1) SCM and procurement, (2) procurement in health care and (3) targeted searches of procurement and SCM in the defence and pharmaceutical industries. Here we summarise our principal approach; further details are described in Appendix 1.

1. General SCM and procurement As a first step, we undertook a systematic search for studies that described any initiatives and mechanisms in procurement and SCM across any sector. We searched MEDLINE, Cumulative Index to Nursing and Allied Health Literature (CINAHL), PsycINFO, Academic Search Complete, Social Sciences Abstracts, Military and Government Collection, EconLit and Business Source Complete from January 2006 to November 2013.

2. Procurement in health care In a second step, we conducted a further targeted search of studies of procurement in health care, using MEDLINE only. We used medical subject headings, with the search extending from 2007 to 2013.

3. Procurement and SCM in the defence and pharmaceutical industries We conducted targeted searches of studies of procurement and SCM in the defence industry, using Google Scholar, for the period 2008–13, and procurement and SCM in the pharmaceutical industries, using Google Scholar, Web of Science and Business Source Complete, for the period 2006–13.

The first search was the most extensive with respect to databases and date range as this was the main source for evidence, whereas the two additional searches were more targeted towards the nature of studies in each particular field.

Inclusion and exclusion criteria

Table 3 summarises the inclusion and exclusion criteria which we applied when selecting studies for review.

Study selection

To ensure consistency in study selection, three reviewers screened the same 200 articles, each using the inclusion and exclusion criteria described in Table 3. Disagreements and discrepancies were resolved by discussion or involvement of a fourth reviewer where necessary. The full list of records (n = 13,191) was then divided between three reviewers for further screening according to the inclusion and exclusion criteria.

Data extraction

Data from studies identified as eligible for inclusion in the review were extracted into a spreadsheet template. We extracted information on context, study design and objective(s), methodological approach, reported outcomes and identified limitations. Data extraction was undertaken by three researchers. Consistency of data extraction across reviewers was checked through duplicate extraction of a random sample of studies (n = 100) by two reviewers independently. Disagreements and discrepancies were resolved by discussion or involvement of a third reviewer where necessary. Given the wide range of types of studies, to aid with the extraction and reporting of the review we have utilised the Context, Interventions, Mechanisms, Outcomes (CIMO) extraction approach, a framework used in management and organisational settings. 41 Details of all studies selected for review are included in Appendix 2.

Quality assessment of studies

Given the heterogeneity of study designs considered in this review, we did not apply a formal quality rating system, such as the Grades of Recommendation, Assessment, Development and Evaluation (GRADE) system for evaluating the quality of evidence for reported outcomes, and typically used for health-care randomised trials. Initial scanning of identified records revealed that many studies were theoretical in nature and did not present empirical data, or that recommendations were not tested in practice. Thus, as a pragmatic approach we applied the following criteria to assess the quality of the studies:

• The research question or aim of the study is clearly stated.

• The approach/mechanism/intervention is clearly defined.

• The study design is rigorous and clearly reported.

• The results are clearly reported.

Study analysis

Data were analysed drawing on the principles of narrative synthesis, which has been recommended as the most appropriate approach for analysing diverse evidence. 42 Building on tabular presentation of findings as described above (see Data extraction), we first grouped data according to emerging themes and particular areas of learning, including, but not limited to, the following:

1. types of ‘end product’: service, product or product–service system

2. types of improvement gained: efficiency, effectiveness, other forms of optimisation and streamlining

3. types of cost savings: transaction costs, items costs, other optimisation costs

4. types of opportunities for innovation: e-procurement, collaborative agreements

5. types of outcomes achieved: purchaser experience, wider economic impacts.

In a further step, we organised our learning from the literature into a logic model format as a means to group the emerging themes and areas of learning into a hypothetical NHS context (Figure 1). In line with the established approach to logic modelling, we distinguished inputs into an organisation (i.e. the context and environment in which purchases are made), the processes or mechanisms used for purchasing, and purchasing outputs and outcomes emerging. In line with our inclusion criteria, we sought to identify outcomes that were (or could potentially be) associated with cost savings, efficiencies and general effectiveness in achieving a given organisation’s aims. We return to a further development of this model, containing the emerging themes from the study, in the discussion (see Chapter 5, Although the evidence remains limited, it is possible to draw some general lessons).

FIGURE 1.

Framework for analysis of studies.

‘Green’ supply chain

Our review identified seven studies45,46,49,50,59,61,68 which specifically reported on initiatives to create a more environmentally sustainable, or ‘green’ supply chain as referred to in this section (other studies also alluded to this theme but are not referenced here). For example, one meta-analysis examined the relationship between environmental SCM (environmental effort targeted towards creation, development and/or delivery of a product to end user) and the firm’s operational performance. 61 It found that, overall, environmental supply chain practices were associated with improved firm performance. An association between ‘green’ supply chain practices and organisational performance was also demonstrated by Zhu et al. (2008),68 in a study surveying automotive organisations in the UK (n = 39) and China (n = 89). It found the association to be positive, if only weakly so, with impacts of the use of environmental SCM practices on performance being due to decreased fines for environmental accidents (UK companies) and an increased volume of goods delivered on time.

Within these articles we found some examples of primary studies. One concerned the tourism sector in the UK and in Europe more widely, and sought to identify examples of good practice to promote sustainability across the whole supply chain among tour operators. 59 Drawing on interviews and document analyses across tour operators (n = 18), tour operators’ associations (n = 3) and non-governmental organisations engaged in tourism (n = 4), the authors documented changes in the supply chain such as the development of standards and assessments, environmental auditing and renewable energy use. These developments were reported to have resulted in financial gains to the organisation such as increased revenues and reduced costs, as well as non-financial gains such as improved brand reputation, staff morale and retention, long-term business relationships with suppliers, retention of clients, operational efficiency and risk management, as well as staying ahead of legislative requirements and protection of the core assets of the business. 59

Azevedo et al. (2011)45 reported one example in the automotive industry in Portugal, finding that, in one firm, operational costs and business wastage decreased following the introduction of one ‘green’ intervention (reusable packaging). Cost savings may also arise from the financial incentives for switching to green product choices; an article in a professional journal reported on various trusts switching to more sustainable practices which in some cases resulted in cost savings of around 30%. 49

In summary, evidence reviewed here suggests that a ‘green’ supply chain can increase staff morale and organisational reputation. There is also some evidence to suggest that moving to a green supply chain may be associated with cost savings. However, studies frequently failed to provide a detailed account of how such reported savings were arrived at and often relied on savings ‘implied’ as a result of incentives for purchasing more sustainable products.

Collective purchasing

We identified one study which reported empirical evidence of cost savings associated with the use of collective purchasing, by which organisations come together to aggregate purchasing. Birk (2009)48 reported on a case of 14 hospitals in the USA that formed a purchasing ‘coalition’ which secured volume purchases, committing to a specific volume as a single unit. Reported savings achieved through group purchases were in the region of 10–25%, with a reported total of US$11M of capital acquisitions saved across the hospitals participating in the coalition; one hospital was reported to have saved just over US$1M on medical supplies in a year.

The limited empirical evidence of collective purchasing as identified in the evidence review does not permit drawing general conclusions about the likely effectiveness and potential of such practices. Interviews with a small number of key informants working in or with the NHS in England highlighted that such practices formed a necessary requirement for effective procurement, mainly because of a perceived lack of specialised purchasing skills within individual NHS trusts (interviewee 3, private sector). There was recognition among interviewees that collective purchasing can help to negotiate lower product prices. This was seen to be particularly important for smaller facilities with limited ‘purchasing power’, whereas a larger hospital ‘with the international reputation gets the better deal than the small little hospital’ (interviewee 5, NHS); however, the usefulness of such collaborations would depend on the nature of products to be purchased:

Collaborative procurements hubs are probably more powerful than they’ve ever been . . . and that can only be beneficial. But that’s only for small medical devices, mainly disposables . . . Really the ones we’re talking about are the electro-medical capital equipment.

Interviewee 5, NHS

In this particular case it was suggested that direct negotiation with suppliers on larger equipment remained the preferable option for services providers.

Overall, the evidence on the effectiveness of collective purchasing identified in our study is limited and it is difficult to predict whether or not collective purchasing would be a successful strategy based solely on evidence from elsewhere. Therefore, although we note that the use of collective purchasing may have the potential to increase the purchasing ‘power’ of service providers through strengthening their position in price negotiations, further empirical evidence is needed to assess the extent to which this is effective, the specific roles of market size and volume of purchases, and the further implications for the supply chain as a whole.

Supply chain integration, alignment and quality improvement

We identified 10 studies44,47,51,55,60,63–67 which discussed the role of supply chain improvement in the context of streamlining and integration and quality improvement as a means to improve organisational performance (other studies also alluded to this theme but are not referenced here). The notion of integration was generally discussed with reference to aligning general corporate strategies and priorities among different organisations within a supply chain, and measures of success included enhanced profitability and customer satisfaction,44 as well as increased revenue. 51 The importance of strategic alignment was also discussed in the context of outsourcing of purchasing functions or other functions within an organisation. For example, one study of the manufacturing sector reported that companies whose drivers and motivations were better aligned (in this instance, the drivers for outsourcing were compared with its general competitive priorities) demonstrated better supply chain performance. 63 One study specifically focused on quality improvement measures in relation to SCM. Carpenter (2008)52 surveyed the implementation and benefits of quality management initiatives within their materials management departments across 710 health-care organisations, of which 58% were reported to have implemented a defined quality improvement programme. Initiatives included programmes such as Six Sigma, lean and rapid cycle improvement, all within their materials supplies departments. Reported outcomes included reduced waste or cost (mentioned by 55% of respondents), improved patient satisfaction (49%), increased communication (44%), reduced hospital-associated infections (44%) and improved overall staff satisfaction (43%).

Stakeholder engagement

Stakeholder engagement, although alluded to as an important factor across sectors, emerged as a particular theme within the procurement literature in health care. Evidence identified for this review drew mostly on single case studies, with three studies70,72,88 reporting on four case studies located in the health-care sector in the USA and one in Canada. For example, Williams (2008)87 reported on experiences in one hospital in North Carolina which created teams of supply chain personnel, finance professionals and clinicians to inform procurement decisions. Teams reviewed data on types of products used in departments, and consulted and negotiated reduced prices with suppliers. This resulted in the standardisation of some products, which was interpreted by the authors as promoting patient safety through the use of ‘similar’ devices across the hospital. Williams (2008)87 also reported on experiences of a group of 20 hospitals in Arizona operating under the same health-care provider, where finance staff and clinicians together developed a pricing model which set a fixed price for physician preference items and was informed on a comparison of prices in the national market. This was reported to have saved the provider more than US$3M per year in cardiology, and US$13M overall. A proportion of the cost savings was invested in new equipment and technology, hypothesised to have sustained engagement of physicians. A related example of clinician engagement was reported by Kehoe (2006). 88 Inspired by SCM systems used by local groceries stores, the executive team of a financially challenged group of hospitals in Pennsylvania combined end-users, such as chief nurses, pharmacists and materials management staff, with engagement and IT investment to improve organisational performance through streamlining, standardisation and overall better materials management. The collaboration of clinicians was reported to have enabled the implementation of new processes and contributed to the success of the initiative.

An analysis of stakeholder engagement in one hospital in Canada was reported to have led to a 34% reduction in 1 year in direct spending on operating room supplies, and a reduction of overall theatre supply costs by 42% over 3 years. 72 This was achieved through involving managers, physicians and nurses in streamlining purchasing and taking a more active part in storing information on supplies and inventory to inform new purchases. Finally, Barlow (2006)70 reported on experiences of two hospitals which invested in improved materials management, while also strengthening clinician engagement in procurement decisions. This was reported to have resulted in a new contract for surgical gloves, which in turn was linked to a reduction in nosocomial infections, estimated to have equated to a total of US$1.5M in savings in the two hospitals combined.

Stakeholder engagement was also identified as an important factor in effective procurement decision-making in our exploratory interviews with key informants. In addition to the value placed upon engaging a wider range of stakeholders in itself, engaging clinicians in particular in the purchasing process was felt to lead to better information about the requirements of the purchase, as well as knowledge of other products on the market, which may lead to a more informed negotiation position for a hospital:

Clinician engagement in procurement decisions is key. They have to be involved. One of the things we’re trying to do . . . is actually set up focus groups. So we have focus groups for various types of equipment. But the focus group is involved in various types of procurement elements in their categories of equipment. And it’s bringing in expert clinicians and nursing staff in those areas to advise and to provide input into that process.

Interviewee 5, NHS

In summary, evidence reviewed here suggests that team collaboration and the engagement of practitioners such as clinicians in the health-care sector may have a core role in the procurement process. Selected examples of stakeholder engagement in health-care procurement in North America point to the notion of ‘engagement’ as good practice. However, the empirical evidence demonstrating the value of engagement with regard to overall organisational performance remains weak.

Capability

Strong professional capabilities in the purchasing function were identified as important by a small number of studies, with supportive evidence based on studies that were of a hypotheses-testing nature rather than reporting on interventions in practice. One study conducted in Australia aimed to examine the skills and knowledge required for logistics professionals to be successful in their job, and the perceived importance of these skills. 85 Using a survey of 1300 members of the Chartered Institute of Logistics and Transport in Australia, the five most important skills and knowledge (in order), as perceived by respondents, were personal integrity, managing client relationships, problem-solving ability, cost control and ability to plan. 85 The study included logistics professionals only, and the response rate was low (11%). However, the authors suggested that these skills would form an important requirement for anyone working in a more general capacity in a purchasing function. One other study in the USA investigated the degree to which personality factors affected customer orientation in logistics personnel across a range of industries. 81 It found that traits such as openness to experience, agreeableness and conscientiousness were significantly and positively related to customer orientation.

The role of capabilities and knowledge that can be linked to enhancing purchasing functions was, in part, reflected in the interviews with key informants for this study, although the nature of these capabilities was not necessarily specified:

At the moment procurement is pretty fragmented in the NHS . . . and there’s no way that most trusts can afford to employ the expertise that they need, even if it was available, which . . . it isn’t . . . There’s a lack of understanding of procurements by chief executives, even though it’s a huge element of NHS costs . . .

Interviewee 4, private sector

Respondents attributed perceived failure to adopt effective procurement models and approaches to a range of external and contextual factors, highlighting a need to ‘understand the role of procurement and what it can do’ (interviewee 2, private sector).

Capabilities required in order to undertake effective purchasing can be acquired through training and experience, although there may be options to transfer purchasing functions to other organisations in the form of outsourcing. One review synthesised the related evidence in the health-care sector. 62 It found that outsourcing purchasing functions resulted in general improvements such as improved quality services by the outsourcing companies (in Germany), some service standardisation benefits (UK) and cost reduction (Australia). However, it also highlighted the potential risks that may be involved, such as cultural change, and the need for monitoring of the outsourced suppliers.

Strengthening capability can also be interpreted more broadly in the sense of creating an ‘enabling’ environment to enhance performance. For example, Ciliberti et al. (2011)74 reported on case studies of four companies in Italy and the Netherlands, each looking at the way they had implemented a specific code of conduct and the degree to which staff adhered to the codes, using interviews, observation and document review. They found that where staff followed codes of conduct and generally accepted them, this was associated with improved communication flows, improved supplier selection processes and solving moral hazard issues.

Relationships with suppliers

A small number of studies (n = 4)71,73,78,84 were concerned with the relationships between purchasers and suppliers. We note that these studies did not link such relationships directly to outcomes. One study surveyed professionals working in conference and event management in the UK. 78 Venue managers reported having long-term supplier relationships, and highlighted the non-financial benefits that can accrue from these, such as mutual trust and good working relationships. Reported benefits included consistency, responsiveness and flexibility in service delivery. Similarly, Barnes and Liao (2012),71 reporting on a survey of around 5150 companies in the manufacturing sector in the USA (response rate 69%), pointed to a positive relationship between a company’s investment in strategic partnership, such as long-term relationships, and its overall performance, although details of how this was implemented were not described.

Conversely, key informants interviewed for this study suggested problems with engaging with suppliers, with reported challenges ranging from a perceived inability to involve small enterprises because of existing NHS framework contracts and fragmentation to ‘disincentive[s] to purchase collaboratively’ (interviewee 1, private sector). One study in the consumer goods industry also highlights the role of potential cultural conflicts, reporting that high-performing supply chain organisations had significantly different cultural profiles to those of their buyers compared with low-performing organisations. 73 The expense of equipment maintenance was highlighted by one NHS respondent, who suggested that lower maintenance cost would require long-term relationships with particular suppliers to ensure lower maintenance and serving contracts, while highlighting how this might be difficult to achieve in situations where suppliers are primarily selected on price (interviewee 5, NHS).

Overall, working relationships with suppliers appear to be important with regard to procurement by means of improved information sharing and shared values, which may support negotiation and contracting functions.

Electronic purchasing

By automating purchasing functions, such as through online purchases and software for internal use, organisations may benefit from more efficient inventory control, time savings and cost savings. For instance, evidence from one case study of a construction company in Greece suggested that use of a customised web-based enterprise resource planning system to manage enterprise process event flow may lead to increased efficiency and control. 112 The authors reported a substantial cost reduction in all areas affected by this new system, but they did not quantify the size of the estimated savings. The types of efficiencies achieved included avoiding paperwork; streamlining personnel-related tasks (e.g. timesheets); flexible and efficient resource planning; increased productivity; increased customer satisfaction; increased sales department efficiency through the better use of sales data; and facilitated communication such as through access to real-time data, documents and reports. The authors also reported a reduction of up to 40% in project execution times. 112 Bhakoo and Chan (2011)92 reported on a longitudinal case study of the implementation of an e-business system in a health-care supply chain (manufacturer to pharmacy) in Australia, using direct participant observation, document and archive analysis and interviews. Before project implementation, pharmacy orders were issued via telephone and fax and goods data entered manually into a central computer before being issued to other hospital areas. The electronic ordering system resulted in a reported reduction of 20% in the time taken to receive goods (from 51 minutes to 40 minutes). There was also a reported decrease in incorrect deliveries, from 8% to 3.5% per order, because of improved labelling; however, there was an increase in packing time for the supplier and time taken to label each carton.

One study reported on four examples of electronic procurement in industrial settings. 111 Reported benefits included reduced supplier numbers through better information in a large UK manufacturing business, improved productivity in a European telecommunication firm, and buyers freed up from transactional work to focus on strategic issues in both a European energy firm and a chemicals and agricultural firm. 111 The types of outcomes reported from one hospital setting in the USA which implemented automated supply chains included increased time with patients due to the release of staff time, as the automated system made more aggregated purchasing data available to help them choose from different products. 108 We further note that through automating purchases, an organisation may be able to access more suppliers. For example, the implementation of an automated supply chain information system within a hospital in the USA led to a rise in the number of suppliers the hospital engaged with, from 7 to 43. 105

These examples point to the potential for time savings and efficiencies as a result of automating processes, although the nature of the empirical evidence remains weak. The adoption of automated supply chain processes can lead to efficiencies, such as reducing time spent by buyers to verify purchase orders,100 but outcomes such as these are seldom contrasted with the investment costs required to set up these systems to allow for a full cost–benefit analysis. 105

Stock management and inventory

Stock management and inventory was an area which was most frequently studied in relation to procurement in the health-care sector in particular, and our review identified a series of small case studies based in hospitals which reported on its roles for organisational performance. Examples highlighted the design and implementation of different stock management systems which tend to rely on the adoption of new technology and the use of RFID.

Raviprakash et al. (2009)110 reported on the use of RFID technology to locate medical equipment and mobile assets within wards in health-care settings in the USA. The authors estimated that the providers may have achieved a return of $200,000 per year over the investment into the RFID system, including maintenance costs, while excluding staff productivity gains. In addition, nursing staff were estimated to have achieved a time saving of about 30 minutes per nurse shift, implying freed-up time for patient care.

In the Stakeholder engagement section we reported on one study that actively involved clinicians in inventory management in a Canadian hospital, which was supported by an automated database. 72 The analysis reported substantial savings on purchasing and supply costs, although it is difficult to attribute reported savings – for example a reduction in overall operating theatre supply costs of 42% over 3 years – to inventory management as such, because of the multiple components involved in the intervention. McHugh (2005)106 reported on the experiences of implementing a specially designed computerised system to actively manage a clinical laboratory inventory in one medical centre in California, USA. This was reported to have led to a reduction in overstock (reduced wastage, with fewer products becoming outdated before use) and reduced paperwork for clinical laboratory scientists, due to enhanced management of the supply chain. The study also reported that there were net savings in operating expenses but it did not quantify the volume of savings.

Similar findings were reported by Persona et al. (2008)109 following the implementation of ‘just-in-time’ techniques in two hospitals in Italy, designed to streamline stock management. Though it was not clear how data were collected, the reported impacts of the new approaches to stock management included savings of up to €21,000 in one hospital in the first year of testing, attributed to reductions in the quantities of stored products, out-of-date medicines and order errors. However, as in many of the examples identified in this review, the initial investment costs of implementing the new systems were not explicitly reported. One of the few examples of work that sought to take a broader cost perspective was a study by Ygal et al. (2010),115 who sought to estimate the impact of RFID tracking technology used to automatically replenish medical supplies in nursing units in a Canadian setting, using interviews, observations and secondary routine data. The authors estimated the time savings accrued to nurses and auxiliary staff to be in the region of CA$153,883 and CA$589,424 per annum, respectively, through the transfer of tasks to store personnel, and a reduction of 67% in the walking distance due to reorganisation of storage location. This equated to a saving of CA$3.1M (or 7.5 minutes per day per nurse and working shift). However, these estimates were arrived at through modelling rather than actual cost audits.

The examples of stock management and inventory reported on here reflect the wider literature on data management, as they allude to the automation of processes. This overall theme was also reflected on in our interviews with key informants, who suggested that technology solutions would play an important role in improving the usefulness of data and gathering important information that the NHS cannot currently access:

[In] the vast majority of cases, [each trust] has created its own catalogue if it’s had a catalogue at all. So master data, you’ve got to be joking. If you’re lucky, some kind of data exists in some trust. But that won’t be the same set of data as exists in another trust . . . The NHS has no idea how many items it buys of anything.

Interviewee 3, private sector

Interviewees highlighted best practices from other companies and stressed the importance of what they referred to as ‘master data’. Currently, barcodes are not routinely used in the NHS to record intake of products and this was reported to lead to problems such as cancelled operations due to unknown inventory quantities.

However, interviewees also noted that the reality of managing data about existing supplies, as well as old and new purchases, can be challenging given the existence of multiple purchasing systems for different types of supplies in each hospital, such as for pharmaceuticals, devices and equipment, and other types of non-clinical supplies: ‘Some trusts have seven different purchasing systems. Like any other organisation, they should have one purchasing system . . . So we do seven different transformations for seven different systems. But we don’t mind, it’s just difficult’ (interviewee 3, private sector).

Better data management can support the purchasing function by providing a history of purchasing information and so enabling negotiations with suppliers, as well as supporting the inventory and stock management functions. The available evidence suggests that techniques that lead to improved stock management and tracking have the potential to save costs, in part because they help to reduce waste and, in the health-care field in particular, enhance the quality and safety of patient care by ensuring stock is available when needed and on time.

Benchmarking and price comparison

A third concept emerging from the literature is related to what we have generally identified as benchmarking and price comparison, and arose mainly from studies set in the health-care sector. However, as noted earlier, the concepts identified here overlap and as we will see below, the usefulness of benchmarking and price comparison will be strongly dependent on data management systems.

Information that enables benchmarking and price comparisons may place purchasers in a better position to negotiate prices. For example, one study in the USA reported on interviews with hospital executives on their price benchmarking strategies. 97 Staff in one hospital compared supply prices with those of other hospitals, and then used this information to negotiate a 13% volume pricing discount, resulting in a reported reduction in costs of $1M. One other hospital reported having negotiated pricing on hip and knee implants, which result in reported yearly savings of $113,000.

The potential usefulness of benchmarking and price comparison was also highlighted by two NHS stakeholders interviewed in our study, who pointed to large differences in prices paid for the same devices and consumables across hospitals in the NHS. They suggested that greater transparency by suppliers providing prices to the NHS would allow hospitals to more accurately compare products that are available on the market.

Main features of the New Zealand health system

Health care in New Zealand is financed largely through public sources, mainly general taxation (74.9% in 2011) and the employment-based accident compensation scheme (7.8%). 117 The remaining 17.3% is funded from private sources including private health insurance and out-of-pocket payments (10.9%); in 2011, approximately 31% of the New Zealand population (of around 4.3 million) had some private health insurance. 118 In 2011, national health expenditure was 10.3% of gross domestic product (GDP), which was higher than the Organisation for Economic Co-operation and Development (OECD) average of 9.3% (UK 9.4%), although remaining lower in terms of per capita spending, at US$3182 compared with an average spend across the OECD of US$3339 (UK US$3405). 117 Compared with 2000, health expenditure in New Zealand grew by 5.1% per year in real terms (OECD 4.1%; UK 4.5%). 117

Since 2000, with the introduction of the New Zealand Health and Disability Act, responsibility for the organisation of publicly funded health services has been devolved to 20 geographically defined DHBs. DHBs administer about 75% of public funding and they are responsible for planning, managing and purchasing publicly funded health and long-term care services for the population within their region. 119 DHBs are governed by a board whose members are elected by popular vote or appointed by the Minister of Health; they are overseen by the newly established National Health Board (see Ministerial review of New Zealand health system performance) and are required to report progress regularly against a set of performance measures and other accountability requirements set out in a nationwide service framework. 120

District health boards operate government-owned hospitals, health centres and community services, and also purchase some (mostly elective) services from private hospitals, and long-stay residential and community-based services from non-governmental organisations. 121 They purchase a significant proportion of primary care services through contracts with primary health organisations (PHOs), which were introduced as non-statutory, not-for-profit bodies following the 2001 Primary Health Care Strategy. 119 PHOs organise and manage publicly funded primary care, and subsidise low-cost access to general practitioner (GP) services, covering over 95% of the population. 122 They bring together doctors, nurses and other health professionals providing care in the community, through either employed staff or affiliated provider organisations and individual general practices; enrolment is voluntary for patients. The number of PHOs was consolidated through merger over the years, from an initial 80 to 31 by the end of 2011. 123

Health-care provision is both public and private. Specialists in public hospitals are paid a salary whereas GPs are usually independent, self-employed providers. GPs act as gatekeepers to secondary care; residents are free to choose any GP.

Ministerial review of New Zealand health system performance

In 2008, amid concerns about the future direction of health-care provision and financial sustainability, and a new commitment to create more efficient and accessible public health services, the newly elected government commissioned a Ministerial Review Group to review the performance, quality and sustainability of the New Zealand health system. 29,121 In its 2009 report Meeting the Challenge the group recommended (re)centralising some of the planning and purchasing functions. 124 The overall direction was to reduce bureaucracy, including reducing duplication of ‘back office’ functions of DHBs and PHOs and an aim to ‘reduce waste and inefficiencies within the health system’. 29 The group also advised the integration of different aspects of planning that had so far been undertaken separately and the development of a joint approach for health services, workforce, technology and capital planning across the entire public system. 124

Among the main changes introduced by the New Zealand government in response to more than 170 recommendations issued by the review125 was the establishment of the aforementioned National Health Board to oversee and guide planning at district and supradistrict levels. Appointed by the Minister of Health and supported by a business unit within the Ministry, the National Health Board is responsible for overseeing operational functions that were formerly the responsibility of the Ministry of Health. These include the funding, monitoring and planning of DHBs, as well as the planning and funding of designated national services. 126 Further significant changes included the creation of a Shared Services Establishment Board, which was to become Health Benefits Limited (HBL) (see next subsection) and a Quality and Safety Commission. 125 Furthermore, the National Health Committee, an independent statutory body, was reconfigured to provide advice to the Ministry of Health on new and existing health technologies in the health and disability sector. 127

A new central procurement agency: Health Benefits Limited

As part of its review, the Ministerial Review Group noted how New Zealand’s Pharmaceutical Management Agency (Pharmac) had contained pharmaceutical cost growth and concluded that a similar approach should be adapted to ‘other non-wage costs in back office areas that all DHBs have in common, in order to free up resources for front-line care’. 124 It thus proposed the creation of a new national procurement agency or ‘Pharmac-like national shared service agency with a mandate to manage the assessment, standardisation, management, purchasing, and/or SCM of any of the common back office functions of DHBs that are referred to it by the Minister of Health’. 124

In response, in 2010 the government established HBL as a shared services organisation for the District Health Boards. 128 It is a standalone Crown-owned company, owned by the Ministers of Health and Finance, whose purpose is to ‘help District Health Boards (DHBs) save money by reducing their administrative, support and procurement costs’129 by means of facilitating and leading national initiatives and managing the related implementation programme. 130 Overall, HBL has been tasked with delivering a total of NZ$700M (around £360M) in gross savings to the DHBs between 2010 and 2015. 131 It is anticipated that these savings will arise from cost reductions, cost avoidance, operational efficiencies achieved in administrative and support services or non-administrative areas, and sharing of good practice in administrative and support services. 130

Assessment of the changes in procurement in New Zealand

In its first 2 years of operation, HBL was reported to have achieved cumulative savings of NZ$114.6M (≈ £60M); these were largely attributed to collective procurement initiatives. 133 Forecasts estimate further gross annual savings from around NZ$100M in 2012–13 to NZ$150M per annum in subsequent years, adding to a cumulative gross saving of NZ$795M by 2015–16 (≈ £415M). 129

The overall impacts of the establishment of HBL and the wider changes in the system towards the intended development of a sustainable health system in New Zealand have yet to be demonstrated (NZ KI). The performance of HBL as such can be assessed against the objectives it set out to achieve, that is, the extent by which HBL succeeded in contributing to gross benefits over the 5-year period after its establishment, with defined long-term outcomes providing a further benchmark for measuring success. 129 Important achievements can be seen in centralising the procurement function. Concerns among suppliers that the centralised approach might stifle innovation have been countered by the argument that the processes put in place by HBL have provided for a clear framework for suppliers to work with (NZ KI).

However, challenges remain. For example, one of the key barriers to strengthening procurement in the health sector in New Zealand, in addition to a fragmented approach to procurement activities, has been a perceived lack in procurement capacity and capability (NZ KI). There has been an interest in centralising leadership by bringing together procurement leads into one group to build capacity and optimise training, but such a move has an impact on positions locally.

When interpreting the New Zealand experience, it is important to consider that with a population of 4.3 million the New Zealand market is small. For comparison, the average population size overseen by strategic health authorities in England before the 2012 health reform was around 5 million. 140 The overall bargaining power in New Zealand is thus small compared with larger markets, which are able to leverage size vis-à-vis negotiating power (NZ KI). A key feature of the New Zealand system can be seen to lie in a perceived culture of working together for the benefit of the population (NZ KI), which may facilitate introducing system changes that may be less acceptable elsewhere.

Background

Group purchasing organisations in the health-care sector act as ‘purchasing intermediaries’ that negotiate contracts between health-care providers and manufacturers, distributors and other suppliers of a range of medical goods and services. 141 Through pooling the purchases of these products for their customers, GPOs can negotiate lower prices from suppliers, which may result in cost savings for health-care providers. 142

Group purchasing organisations started emerging in the USA in the late 1950s. 143 According to the Healthcare Supply Chain Association, which represents 14 GPOs in the USA, about 98% of US hospitals use GPOs to purchase products, on average between two and four GPOs per facility. 144 It is estimated that GPO contracts account for just over 70% of non-labour hospital purchases. 144,145 Although a large number of GPOs operate in the USA (> 600), the market is concentrated in a small number, with six of the largest national GPOs accounting for almost 90% of all hospital purchases. 146

During the early 2000s, GPOs became subject to congressional scrutiny because of concerns about potential anticompetitive practices, including the charging of fees to manufacturers by GPOs, raising questions about conflicts of interest, among other things. 141,143 In response, in 2005 nine GPOs established the Healthcare Group Purchasing Industry Initiative to promote best practices and public accountability among its member GPOs. 146,147 Although this effectively operates as a self-regulatory body, the government has retained federal oversight. 146

Empirical evidence of the impact of GPOs on pricing for hospitals in the USA is limited. 142 Studies that are available suggest that GPOs appear to reduce health-care costs by lowering product prices,148,149 and also reduce transaction costs through commonly negotiated contracts. 148

Group purchasing in European countries

Group purchasing in the publicly funded health-care sector has become an increasingly important feature in some European health systems from the late 1990s onwards, in response to a perceived need to reduce fragmentation, inefficiencies and lack of transparency in procurement activities; examples include England, France, Germany and, more recently, Italy. 11 The nature of group purchasing differs across countries, however, with varying involvement of national or regional public agencies in the co-ordination or oversight of procurement activities in the publicly funded system. 150 For example, in Italy collaborative procurement associations or other forms of collaboration have been set up at regional level,28 and the national procurement agency is tasked with the co-ordination of a network of regional central purchasing bodies. 151 In France, public sector reforms in the mid-2000s have encouraged the formation of nationally and regionally grouped procurement consortia, including in the health-care sector,11 with GPOs playing an increasingly important role in the government’s efforts to strengthen public procurement, as we shall see in the next subsection.

In Germany, about 80% of hospitals use GPOs, and the volume of non-labour hospital spending processed through GPOs increased from an average of around 20% in 2000 to 42% in 2010. 152 The organisational structure, remit and scope of GPOs in Germany varies, ranging from hospitals jointly co-ordinating their procurement to non-binding and binding purchasing companies. 153 The main focus of hospitals joining purchasing associations or organisations has been on cost savings through custom contracting, that is, contract and price negotiations,154 although more recently GPOs have begun to offer additional services such as IT, process consulting and logistics services. 153 It has been estimated that in 2010, hospitals in Germany made cost savings in the region of €4B through the use of GPOs. 152 GPOs are also playing an increasingly important role in Austria and Switzerland;152 there are indications of an increasing consolidation and Europeanisation of GPOs, with, for example, Prospitalia, one of Germany’s leading GPOs, which has established subsidiaries in Austria and the Netherlands, also foreseeing the formation of so-called ‘super GPOs’ across the European health sector. 155

Group purchasing in European countries: the experience in France

The French health system is based on statutory health insurance (SHI) and provides all legal residents with health coverage, as per the 1999 Universal Health Coverage Act (CMU Act). 156 In 2011, SHI accounted for 73.1% of health expenditure, complemented by taxation (3.6%), out-of-pocket payments (7.5%) and private health insurance (15.7%). 117 About 95% of the population hold complementary private health insurance to cover user charges and/or excluded services (e.g. psychologists, dieticians). 157 National health expenditure in 2011 was 11.6% of GDP (UK 9.4%), with per capita spending at US$4118 compared with an average spend across the OECD of US$3339 (UK US$3405). 117 Compared with 2000, health expenditure in France grew by 2.5% per year in real terms (OECD 4.1%; UK 4.5%).

Although the Ministry of Health oversees overall health sector planning and provides guidance on health policies, regions have an increasingly important role in health-care governance through regional health agencies [agence régionale de santé (ARS)]. 156 Created in 2010 following the 2009 Hospital, Patients, Health and Territories Act, the ARS are responsible for ensuring that health-care provision meets the needs of the population by improving co-ordination between ambulatory and hospital care and health and social care services, while respecting national health expenditure objectives.

Health services are delivered by public and private providers in ambulatory care and in hospital. GPs mainly work in private practice as self-employed professionals, with around 75% working in health centres or hospitals in addition to their private practice. GPs are reimbursed on a fee-for-services basis, with fees set nationally, based on agreements between professional organisations and the SHI administration. Since 2009, GPs can also enter into individual contracts with the SHI to receive additional payment in compensation for ‘practice improvements’ (pay for performance; from 2011 this was extended to also include specialists). 158 Specialists are paid based on fee for service in both private practice and private hospital settings. Specialists employed in a public hospital receive a salary.

Patients are able to access specialists in hospitals and private practice directly. In 2004, a ‘soft’ form of gatekeeping was introduced to encourage patients to see a GP before visiting a specialist, referred to as ‘preferred doctor’ (médecin traitant). Although a voluntary scheme, there are strong financial incentives for patients to sign up, and more than 85% of the population is registered with a GP. 157

The hospital sector in France

Secondary and tertiary care is provided by a mix of public (including private not-for-profit) hospitals, covering two-thirds of hospital beds, and private for-profit hospitals. 157 There were just under 2700 hospitals in France in 2011. 117 Public hospitals are legally and financially independent but overseen by the state, whereas private not-for-profit hospitals are typically managed by associations, foundations, mutual insurance companies or others. 159 Private for-profit hospitals are civil or commercial enterprises and, increasingly, form large corporations. 157

Hospital care is financed through an activity-based funding system using diagnosis-related groups, which became fully operational in 2008. The system is used to reimburse both public/private not-for-profit and private for-profit hospitals. Private for-profit hospitals have been paid entirely through diagnosis-related groups since 2005. 160 The aforementioned 2009 Hospital, Patients, Health and Territories Act has introduced a number of changes to the governance of public and not-for-profit hospitals, increasing their autonomy and organisational flexibility. 156 This also included devolving executive responsibilities from the administrative board of the hospital (subsequently the monitoring board), which comprises representatives of the state, local authorities, hospital staff, patients and qualified personnel, to the hospital director.

In 2006, the government launched a continuum of initiatives that sought to strengthen efficiency in the French public sector, in particular the introduction of nationally and regionally grouped procurement strategies. 161 As part of this strategy, in 2011 the government introduced the PHARE programme, in an attempt to achieve further efficiency gains in the hospital sector. 162 It is estimated that procurement in the hospital sector amounts to an annual expenditure of €18B, of which around 60% is spent on medical goods and services. 163 The PHARE programme set out to realise ‘smart savings’ by means of providing hospitals with greater flexibility within a financially constrained environment while improving the quality of care provided to patients. 116 The overarching goal was to enable all the levers for effective procurement, including group purchasing or framework contracting, optimising products and services purchased and optimising procurement processes. 116

The programme identified a hospitals’ savings potential of a total of €910M over the period 2012–14. 162 This was to be achieved through the development of a purchasing function within institutions with a unique responsibility for procurement; the development of a regional procurement policy by the ARS; and the development of a national pilot project led by the Directorate General of Health Care Provision at the Ministry of Health [Direction générale de l’offre de soins (DGOS)]. 162

The programme is organised around six areas: (i) procurement performance, which seeks to mobilise all levers and build support through projects (project ARMEN; Box 3) and support for the development of regional markets; (ii) institutional support, including support for the leading 150 institutions in the development of their first shared procurement plans, and development and dissemination of dedicated tools (‘Kit ES’) and training opportunities; (iii) support for regional health agencies (ARS) as moderators and facilitators of strategic procurement at regional level, including for the development and dissemination of specific tools (‘Kit ARS’); (iv) communication including newsletters, websites and meetings of decision-makers; (v) leadership, including monitoring of progress of the programme and possible corrective action; and (vi) launch and management of high-impact cross-cutting projects such as streamlining the procurement process, procurement information systems and supply chain. 162

Group purchasing organisations as key players within the Performance Hospitalière pour des Achats Responsables programme

One of the aims of the PHARE programme was to demonstrate ‘quick wins’ in its efforts to strengthen procurement in the health-care sector. 162 This was to be achieved, in part, through building on and bringing in the expertise of existing operators in the market, identified as the four large public, not-for-profit GPOs: UniHA, UNICANCER, Union des Groupements d’Achats Publics (UGAP) and Réseau des acheteurs hospitaliers d’Ile de France (Résah-idf). In brief, the French public sector procurement co-operative UniHA represents 56 hospitals across France, including 30 university hospitals; it represents half of all public hospital procurement in France. 165 The UNICANCER group brings together the 20 French comprehensive cancer centres and their federation and pools their strategic activities: research, purchasing, human resources, hospital strategy, quality control and information systems. 166 UGAP is the only general public procurement agency in France, which represents public sector organisations in different areas including health care. 167

We here focus in on Résah-idf, which has been tasked by the Ministry of Health, as part of the PHARE programme, to support and co-ordinate the inter-regional network of health-care group purchasers (Alliance Groupements), bringing together over 100 groups, with an annual procurement volume of €8B. 168

Assessment of the Performance Hospitalière pour des Achats Responsables programme

Since the inception of the PHARE programme in 2011, reported progress has included an initial pilot phase that sought to demonstrate potential savings with one region and one facility as well as encouraging key actors to engage in the programme, and a subsequent expansion of the activities to cover a larger number of regions (at least 10) and facilities (20–30). 162 Since June 2012 through 2014, the programme is being rolled out nationally, including an acceleration of exchange of best practices, support for the top 150 institutions in the implementation of their first shared procurement plans and promotion of the creation or consolidation of regional procurement policies. Given the ongoing roll-out of the programme, it is too early to assess its overall impact. Important achievements of the programme could be seen as having placed procurement as a strategic issue on the agenda, and its facilitation of the networking of actors within and across regions (Fr KI). Recent figures from the French Ministry of Health suggest that by mid-2013 savings of approximately €200M had been achieved within the PHARE programme. 163

A key feature of the French approach to public procurement in the health sector is the emphasis on the region as a hub for group purchasing activities. The regional unit, with a procurement volume in the region of €1.2–1.5B, can be seen to provide potential for the demand side to take advantage of savings achieved through collective procurement while allowing sufficient scope for suppliers to engage in the process and guarantee contracts (Fr KI). Elsewhere, there has been concern that increasing centralisation of the procurement function, whether at regional or national level, may have a negative impact on the adoption of innovation in the health-care sector, with the possible exemption of ‘breakthroughs’, because of demand standardisation and reduction of market prices. 150 Related concerns have been raised around the extent to which collective procurement may disadvantage small- and medium-sized enterprises (SMEs), for example in relation to large volumes or administrative requests under public tendering conditions. 150 Empirical evidence regarding whether GPOs promote or hinder innovation is rare and conflicting. Burns and Lee (2008),148 based on a national survey of hospital purchasing groups in the USA, did not find evidence that GPOs exclude new innovative firms from the marketplace, while Hu and Schwartz (2011),149 using modelling techniques, concluded that GPOs reduce manufacturers’ incentives to introduce innovations to existing products. These two findings are not mutually exclusive, however, and highlight the need for further research to understand the impact of GPOs and collective procurement more generally on innovation. 11

Search strategy

The review presented in this report sought to capture evidence in the area of procurement and SCM in a wide range of sectors to identify best practices that may inform practice in the NHS. In order to manage the potentially large number of eligible studies for review within the time frame of this study of 9 months, it was necessary to focus the search terms while keeping the search area as broad as possible. The nature of the subject area also meant that approaches to systematic searches that, in the health field, can draw on medical subject headings, were of limited applicability. This meant that we had to test the performance of free text search terms, in isolation and in combination, to assess the volume of studies of potential relevance. Given the large number of records identified by initial searches we used an iterative process to narrow the searches, working with an information scientist, to arrive at a manageable number of studies for screening and further analysis. Our final search strategy applied a fairly restricted combination of search terms, requiring presence of both ‘supply chain’ and ‘procurement or purchasing or supply chain management’ in title, abstract or subject (see Appendix 1). Although this still yielded a large number of studies (in excess of 12,000), it is possible that this approach missed studies that would have been of relevance for this review.

We further narrowed the review by including studies that reported empirical findings only. However, as empirical work in the field of procurement is limited, we broadened our definition of ‘empirical’ to also include single case studies. Given our broad approach to the search, it is perhaps not surprising that studies identified as eligible for inclusion in the review covered a very wide range of aspects in the fields of procurement and SCM, with few examples of actual interventions in practice, which may simply be a reflection of the field and the methodological approaches that are typically used. It is possible that an alternative approach which had focused on specific themes within the procurement and SCM fields might have identified studies that would have informed specific learning in particular aspects. However, the call for proposals for the present review asked for an overview of the field, and we believe such a broad approach provides a necessary starting point to guide future research that focuses in on specific questions around procurement or SCM to inform learning for health care more broadly and the NHS specifically.

We imposed a limit on the publication date of studies to be considered for review and used the year 2006 as a cut-off, excluding earlier studies. We accept that by restricting the inclusion of potentially eligible work in such a way we may have inadvertently excluded earlier studies that could have provided important lessons to be drawn. The cut-off was chosen for pragmatic reasons, noting that certain aspects such as the use of technology and sustainability have emerged as core themes only since the mid-2000s. The evidence assessment further excluded studies published in languages other than English and those set in low- and middle-income countries. Again, by imposing these exclusion criteria we may have missed potentially important developments in those settings, although we should emphasise that the health care and system infrastructure in low- and middle-income countries is very different from that of high-income settings, likely restricting the transferability of experiences to the NHS context.

Keeping these limitations in mind, we should, however, note that we feel confident that our searches and review have captured themes that we understand to be most pertinent to the field of procurement and SCM as it concerns the health-care sector. This assertion is further supported by the key themes that have emerged from interviews with a small number of stakeholders in England, which tend to mirror the themes identified from the literature.

Furthermore, we believed it to be important to also consider the health-care sector itself as a direct source for both best practice and learning, and we consider the inclusion of health-care-specific studies to be adding value to the final selection in terms of lessons learned of relevance to the NHS. These studies also added an important comparative component to our analysis, in that the concepts analysed from examples in other sectors could be compared with initiatives currently tested within health-care settings.

Finally, in our initial research protocol we proposed undertaking systematic assessments of the experience of procurement and SCM in the health-care sector in other high-income countries. However, as described earlier, following the 2013 publication of the NHS procurement development programme9 we believe that in-depth assessment of country experiences in this field more generally would have added little additional value for learning, given that the NHS procurement programme will be implemented in due course. We therefore focused on two specific examples of countries that have recently also introduced a strategic framework or reform to guide procurement in their respective health-care sectors. We accept that by following this approach we may have unduly limited the potential for learning from elsewhere, although we should note that this component of the work was aimed at complementing the REA rather than presenting a separate study in itself, which would have been outside the scope of this research.

Framework for analysis

We should note that our initial research protocol anticipated a review of the evidence on procurement and SCM in areas other than health care to inform potential lessons for the health-care sector in England by identifying models of good practice. However, on conducting an initial scan of the published evidence it became clear that distinct ‘models’ of procurement and SCM were not present in the literature and in the field overall in a way that would have allowed appraisal of models by sector. Instead, the published literature focused on different approaches to purchasing and SCM practices described for different settings. Against this background we believed it to be useful to adopt a more ‘emergent’ approach by allowing the themes to emerge from the literature itself. In addition, we note that the original protocol presented a framework for analysis which would have allowed us to categorise the learning from both the theoretical models and the individual industry sectors into specific ‘areas for improvement’ for the NHS. However, we also noted that a review of practice in other sectors would require further analyses to demonstrate transferability and applicability to the health-care sector. Firstly, transferability in identifying the types of practices, such as, for example, environmental approaches or group purchasing that could be beneficial to a health-care setting, would have to be analysed through further study. In turn, further analyses would be required to understand the mode of practice that would be applicable to specific areas within health care, for example large or small equipment and devices, consumables and stationery, where different approaches might accrue differential benefit. Such an approach was, however, beyond the scope of this study, although we acknowledge that the use of such a framework may help guide future research. We therefore chose an analytical framework that served to synthesise the themes emerging from the literature in the form of a logic model used to group the emerging themes and areas of learning to a hypothetical NHS context (presented in Chapter 2, Figure 1). However, we caveat that the model was not intended to identify causal links between identified processes and outputs specifically, and was used as a means to summarise the themes only. Its relation to the emerging themes is described further in the discussion of the findings (see Although the evidence remains limited, it is possible to draw some general lessons).

The nature and quality of the evidence of interventions in the fields of procurement and supply chain management is diverse

Thinking within and approaches to the fields of procurement and SCM have evolved considerably during the past decades, and this is reflected by wide variation in scope and domain, definitions, key trends, theories and methodologies seeking to characterise the fields. 173 SCM has been defined as a multicomponent field encompassing the four areas of logistics, marketing channels, purchasing and operations management. 173 Therefore, any review within and of this field will include studies across these areas, along with the wider range of methodologies employed across each. This broad scope is reflected in the literature identified in our initial scoping of the available evidence for this report, which can be categorised as follows:

• hypothesis-testing studies, validated empirically through surveys or interviews

• case studies of current practice

• studies that test a model or theory without presenting empirical evidence

• studies with scholarly contribution only (i.e. setting recommendations for further scholarly research only).

Of these, we only considered studies that presented some form of empirical evidence as eligible for inclusion in the full review. Evidence identified covered a range of sectors and industries, including textile, IT, the automotive industry and manufacturing, alongside the health-care sector.

Overall, the body of empirical work in the fields of procurement and SCM as identified here was limited, both in quantity and quality. At the outset it was challenging to identify examples of good practice given the very theoretical nature of much of the literature in these fields. Studies presenting practice examples tended to be rather weak in terms of methodology, lacking adequate description of methodological design, modes of data collection and strategy for analysis. This observation is not necessarily new, however. For example, Carter et al. (2003),174 in a review of studies published in the Journal of Supply Chain Management, found that nearly 90% of the articles published from 1965 to 1999 consisted of normative literature, methodology reviews and exploratory studies; only a small proportion of studies provided empirical studies involving hypothesis testing. They highlighted the need for systematic assessments of the existing literature as well as rigorously designed inductive studies that would allow for the development of frameworks and testable research propositions. Subsequent work by the same authors further highlighted a possible trend away from the use of empirical data collection (such as surveys and case studies) and corresponding methodologies towards econometric modelling, at the possible expense of empirical approaches. 175 The need for empirical work was also suggested by review work included in our report, such as the study by AlSagheer and Ahli (2011),44 who assessed the evidence base for supply chain integration in relation to business performance. This was recently also highlighted in relation to the broader field of purchasing or commissioning of health services, with Allen (2009)176 noting how lesson drawing for the NHS is challenged by weaknesses in the existing evidence base.

In addition to the relative lack of empirical evidence, our review highlights that where data are presented, studies frequently fail to assess (or describe) the robustness of their methodological approaches when linking interventions with outcomes, such as cost savings or improved performance. Typically, although not always, the review had to rely on single case studies and evaluations of an intervention were seldom presented. Where cost data were presented, analysis frequently did not consider initial investment costs associated with a given new intervention, so it is difficult to arrive at firm conclusions on return on investment and efficiency gains more broadly.

We should note that identifying evidence from the procurement and SCM fields that may be applicable to health care, or any area of the public sector, remains challenging. Thus, a recent literature review on the use of processes developed in the private sector within the public sector showed that 51% of publications focused on ‘lean’ methods (i.e. streamlining processes) and 13% on business process re-engineering, while 35% stated their use in health services. 177 That review had to draw mainly on case studies, and the authors noted that these usually failed to consider the influence of factors such as context, behavioural aspects, organisational design or policy implications. Others have argued that the majority of studies on methodologies such as lean in the public sector tend to lack a suitable comparator or generally a rigorous study design. 178

Although the evidence remains limited, it is possible to draw some general lessons

The evidence review, complemented by interviews with key stakeholders and insights from experiences in other countries, identified three general themes of potential learning: (i) organisation and strategy, (ii) collaboration and relationships, and (iii) materials and information management. We discuss the main insights below.

Overview of key learning from the studies

Building on our framework described in Chapter 2 (see Figure 1), we have organised the key learning emerging from the findings of the study presented in this report in the form of a logic model to illustrate the potential applicability of our findings to a (hypothetical) NHS context (Figure 3). Inputs refer to the knowledge of available product for purchase within an organisation, but also the professional capabilities of those working in purchasing functions, which was reflected in both the evidence review and interviews. Under processes we list the types of approaches identified in this study, such as collaboration and engagement, electronic tagging and information sharing. We further list the types of outputs that arise from these processes, which we extracted from the literature, although we caveat that most of these were reported as narratives and not evaluated after implementation. These outputs could potentially be linked to potential outcomes such as cost savings, general efficiencies (resulting, for instance, from reduced staff administration time) and more effective performance management. We note that this framework guided the analysis of the evidence, serving to illustrate the types of associations between different components as demonstrated by the evidence reviewed. However, as we have highlighted, the quality of the evidence was frequently low and did not permit drawing robust conclusions as to the direction of (implied or empirically verified) links between the processes, outputs and outcomes.

FIGURE 3.

Framework for analysis including identified lessons (themes) from the study.

Implications for the NHS

Although we clearly caveat the challenges in taking lessons from other industries for the health-care sector, we provide some of our observations on selected aspects of the learning gained in terms of their implications for the NHS.

Search terms

The logic links between the different categories were: 1 AND 2.

Search terms should be found in the title, abstract or subject keyword.

Results: n = 12,577.

Search terms (Major MeSH)

The logic links between the different categories were: 1 OR 2.

Search terms should be found in the title, abstract or subject keyword.

Search terms

The logic links between the different categories were: 1 OR 2.

Search terms should be found anywhere in the article.

Search terms

The logic links between the different categories were: 1 OR 2 OR 3.

Search terms should be found anywhere in the article.