Health and Social Care Delivery Research (Oct 2024)
Consequences of how third sector organisations are commissioned in the NHS and local authorities in England: a mixed-methods study
Abstract
Background As a matter of policy, voluntary, community and social enterprises contribute substantially to the English health and care system. Few studies explain how the National Health Service and local authorities commission them, what outputs result, what contexts influence these outcomes and what differentiates this kind of commissioning. Objectives To explain how voluntary, community and social enterprises are commissioned, the consequences, what barriers both parties face and what absorptive capacities they need. Design Observational mixed-methods realist analysis: exploratory scoping, cross-sectional analysis of National Health Service Clinical Commissioning Group spending on voluntary, community and social enterprises, systematic comparison of case studies, action learning. Social prescribing, learning disability support and end-of-life care were tracers. Setting Maximum-variety sample of six English local health and care economies, 2019–23. Participants Commissioning staff; voluntary, community and social enterprise members. Interventions None; observational study. Main outcome measures How the consequences of commissioning compared with the original aims of the commissioners and the voluntary, community and social enterprises: predominantly qualitative (non-measurable) outcomes. Data sources Data sources were: 189 interviews, 58 policy and position papers, 37 items of rapportage, 692,659 Clinical Commissioning Group invoices, 102 Freedom of Information enquiries, 131 survey responses, 18 local project group meetings, 4 national action learning set meetings. Data collected in England during 2019–23. Results Two modes of commissioning operated in parallel. Commodified commissioning relied on creating a principal–agent relationship between commissioner and the voluntary, community and social enterprises, on formal competitive selection (‘procurement’) of providers. Collaborative commissioning relied on ‘embedded’ interorganisational relationships, mutual recognition of resource dependencies, a negotiated division of labour between organisations, and control through persuasion. Commissioners and voluntary, community and social enterprises often worked around the procurement regulations. Both modes were present everywhere but the balance depended inter alia on the number and size of voluntary, community and social enterprises in each locality, their past commissioning experience, the character of the tracer activity, and the level of deprivation and the geographic dispersal of the populations served. The COVID-19 pandemic produced a shift towards collaborative commissioning. Voluntary, community and social enterprises were not always funded at the full cost of their activity. Integrated Care System formation temporarily disrupted local co-commissioning networks but offered a longer-term prospect of greater voluntary, community and social enterprise influence on co-commissioning. To develop absorptive capacity, commissioners needed stronger managerial and communication capabilities, and voluntary, community and social enterprises needed greater capability to evidence what outcomes their proposals would deliver. Limitations Published data quality limited the spending profile accuracy, which did not include local authority commissioning. Case studies did not cover London, and focused on three tracer activities. Absorptive capacity survey was not a random sample. Conclusions The two modes of commissioning sometimes conflicted. Workarounds arose from organisations’ embeddedness and collaboration, which the procurement regulations often disrupted. Commissioning activity at below its full cost appears unsustainable. Future work Spending profiles of local authority commissioning; analysis of commissioning in London and of activities besides the present tracers. Analysis of absorptive capacity and its consequences, adjusting the concept for application to voluntary, community and social enterprises. Comparison with other health systems’ commissioning of voluntary, community and social enterprises. Funding This award was funded by the National Institute for Health and Care Research (NIHR) Health and Social Care Delivery Research programme (NIHR award ref: NIHR128107) and is published in full in Health and Social Care Delivery Research; Vol. 12, No. 39. See the NIHR Funding and Awards website for further award information. Plain language summary National Health Service commissioners and local councils often buy health and care services from voluntary, community and social enterprises. This study aimed to explore how commissioners and voluntary, community and social enterprises worked together and where improvements could be made. We talked to commissioners and voluntary, community and social enterprises in six areas across England and focused on services for learning disabilities, social prescribing and end-of-life care. We analysed National Health Service financial accounts to see how much the National Health Service was spending on services provided by voluntary, community and social enterprises. We surveyed how commissioners and voluntary, community and social enterprises were using information and knowledge to make decisions. We organised events bringing together commissioners and voluntary, community and social enterprises to share knowledge and experience. We found there were two ways commissioners bought services from voluntary, community and social enterprises. One was commodified (a buying-and-selling model), the other collaborative (based on working together). Both were happening at the same time in all the areas of the study, but usually one of them was more present than the other. We saw a general move towards collaboration, but some areas were further along with this than others. Various things helped commissioners and voluntary, community and social enterprises collaborate, such as: paying voluntary, community and social enterprises enough for their services; having people and networks that encouraged others to work with voluntary, community and social enterprises; and including voluntary, community and social enterprises in making decisions about health and care. Commissioners and voluntary, community and social enterprises might therefore lean further towards working together to see how collaborative commissioning can be further developed; how to make contract prices cover voluntary, community and social enterprises’ costs; how to enable longer-term contracts; and how to enable less hurried, more considered ways of renewing contracts (e.g. by putting draft contract specifications out to pre-tender consultation). The new Integrated Care Systems in local areas could encourage all this to happen, but pressures elsewhere in the health and care sector might make it more difficult. Scientific summary Background This study examines how the NHS and local authorities commissioned voluntary, community and social enterprises (VCSEs); some outcomes for commissioners, VCSEs and the health system; and which contexts affected these outcomes. Existing studies describe how the public funding of independent providers occurs through a commissioning cycle of service specification, provider selection, contract-letting and monitoring of the activities actually delivered. The original policy (and theoretical) formulations of this cycle presupposed a clear separation of commissioners from providers. In practice, however, the cycle is partly implemented, and attenuated, by co-commissioning activities, collaborations in which potential providers, citizens and the public contribute at each stage. Commissioning has been explained as, inter alia, a means by which governments continued to exercise governance (not direct control) over independent providers of public-funded services through six main media of power: managerial techniques, negotiated order, discursive control, resource dependencies, provider competition, and juridical control. Each particular combination constitutes a ‘mode of commissioning’. Previous studies have examined the modes of commissioning applied elsewhere, but not to VCSEs, and then usually from the standpoint of policy-makers’ aims and intended service outcomes of commissioning. Fewer studies have explored commissioning from the VCSE standpoint. Some that did reported largely negative consequences for the commissioned VCSEs: a loss of freedom to criticise policy and a ‘degeneration’ of the VCSEs’ democratic internal regimes. Still fewer studies have closely examined the interactions between commissioners and VCSEs as they engage in the commissioning cycle and in co-commissioning. The practical import of these questions is whether the commissioning of health-related VCSEs enables them to supplement the reach of NHS activities and strengthen users’ voice in the health system, or whether it undermines the characteristics of VCSEs which first motivated the commissioning of them. Because commissioning involves two main groups of agents, their interactions involve not just one but two context–mechanism–outcome configurations (CMOCs). In one, the commissioners are the focal actor (as in most realist evaluations); in the other, VCSEs are. The two configurations intersect at the commissioning mechanisms, where the parties interact. Earlier studies suggested that the parties’ absorptive capacity (ACAP) to acquire, assimilate, transform and exploit externally sourced knowledge was an important context affecting what interactions occurred, and to what effect. Objectives This study aimed to produce knowledge about which factors strengthen (or weaken) collaboration between healthcare commissioners and VSCEs, and make commissioning relationships between the commissioners and VCSEs more productive for all. Research questions were: RQ1. How do healthcare commissioners address the task of commissioning VCSEs as service providers, and what barriers do they face? RQ2. What are the consequences for VCSEs of the public bodies commissioning services from them? RQ3. How are VCSEs involved in Clinical Commissioning Group (CCG), local authority and other [e.g. Integrated Care System (ICS), NHS England] commissioning decisions? RQ4. What ACAPs do healthcare commissioners and VCSEs respectively need for enabling VCSEs to be commissioned, and for co-commissioning? Methods This study was a mixed-methods realist analysis of the mechanisms by which English health-related VCSEs are commissioned. To investigate the intersecting CMOCs, we used five main methods: Preliminary scoping work with national-level NHS and VCSE organisations to identify important current developments in this domain and likely data sources, based on interviews and content analysis of policy documents. A cross-sectional profile of CCG spending on VCSEs, which provided data about patterns of VCSE commissioning and was a sampling frame for the three following work packages. We content-analysed 226,138 CCG invoices (for sums over £25k) from 2018 to 2019 to discover the distribution of CCG spending on VCSEs and how it compared with CCG spending on non-VCSE providers. Using findings from the preceding work packages, we drew a sample of six places contrasted by their proportion of spending on VCSEs. We systematically compared case studies of VCSE–commissioner collaboration in formulating local commissioning strategies (‘co-commissioning’) in them (2020–2). Commissioners’ and VCSEs’ aims when engaging in commissioning were examined separately for either side, as were the outcomes relevant to each, but not the structures and activities through which they interacted. A systematic comparison of case studies of the commissioning of VCSEs, using the same study sites, methods and unit of analysis as the preceding work package (late 2020 to summer 2022). Action learning activities (2020–3) in the same sites: Local project reference groups supported the project in each study site and combined as: national action learning workshops a preliminary, exploratory (not randomised sample) survey of ACAP in the study sites local co-researchers who conducted local research projects into the commissioning of VCSEs and whom the research team mentored As tracer studies we examined social prescribing, end-of-life care and support for learning disabilities, on the assumption that this selection gave variety in the scale and number of VCSEs involved, and type of activity (clinical vs. preventive). Framework analyses were used to synthesise the five sets of findings and map them onto the research questions. Results Two modes of commissioning VCSEs existed in parallel: A commodified mode centred on the commissioning cycle, financial dependencies, formalised procurement [a regulatory (i.e. juridically based) practice] and provider competition. Consulted VCSEs were often excluded at either service specification or provider selection stage in the cycle. The financial dependency of VCSEs was a central medium of power for commissioners. Juridical power was in the background but seldom used in practice. The paradigm form of this mechanism was a direct bilateral principal–agent relationship between one commissioner and one VCSE, with the latter wielding less power. This was a commodified mode of commissioning. To reduce their transaction work (costs), commissioners were introducing larger contracts (e.g. with one large ‘lead’ provider subcontracting many smaller VCSEs). Networks for negotiating what activities were commissioned and how they were implemented. Commissioners and VCSEs were both embedded in these networks, which constituted an interorganisational negotiated order among commissioners, among VCSEs, and between commissioners and VCSEs. Persuasion and legitimation were the main discursive media of power, supplemented by relationality (mutual trust) and mutual ‘real-side’ (as opposed to financial) resource dependency. Workarounds such as alliance and lead provider commissioning were important media of managerial power, adopted in order to diminish the expected, and in some cases observed, adverse consequences of commodified commissioning. VCSEs’ commissioning and co-commissioning activities overlapped considerably. The networking mechanisms supplemented and attenuated the quasi-market mechanisms, and indeed were partly intended to. This was a collaborative mode of commissioning. It was not that where one mode of commissioning existed, the other did not; both coexisted and interacted everywhere but the balance between them varied. Certain contexts affected how these mechanisms worked in practice: Local health and care system characteristics: Fiscal constraints upon commissioners constrained how much VCSE activity could be commissioned, compelled commissioners to prioritise VCSE activities that appeared to offer immediate cost savings elsewhere in the health system, and destabilised VCSE income. Diverse ownership mix of provider organisations (e.g. in learning disability services) motivated more cautious, commodified commissioning insofar as commissioners anticipated challenges to their decisions from unsuccessful bidders. Having a small number of stable VCSEs (e.g. in end-of-life care) enabled and necessitated collaborative commissioning. Geographical and historical characteristics: Population deprivation motivated commissioners and VCSEs to respond through collaborative commissioning. Spatial dispersion increased the practical difficulty and costs of, and the number of networks involved in, collaborative commissioning. Co-terminosity of commissioners minimised the number of interorganisational interfaces that the networks had to span. Local policies, ethos and history strongly influenced whether collaboration and trust between commissioners and VCSEs was long-established, and thus how collaborative commissioning could be. The presence and nature of networking spaces, in particular which VCSEs participated, and how much voice they had when they did participate, affected how collaborative co-commissioning could be. Organisational characteristics: ACAP affected commissioners’ ability to know what resources VCSEs could offer, and what VCSEs were aiming for, in becoming involved in commissioning, and vice versa. Commissioners relied on formal sources of information and evidence more than VCSEs tended to. Discursive (translation) gaps between VCSEs and commissioners were widespread. Organisational systems, culture and bureaucracy: the more narrowly procurement regulations were interpreted and implemented, the more commissioning was commodified rather than collaborative. Individuals’ role, discretion and influence: individuals who lobbied for VCSEs to engage, or be engaged, in commissioning and who were boundary-spanners with knowledge of both commissioning and VCSEs facilitated the development of collaborative commissioning, as did ‘maverick’ individuals who devised workarounds when procurement regulations appeared to obstruct the commissioning of VCSEs, especially collaborative commissioning. The nature of VCSE activity: Statutory or non-statutory status: statutory requirements for service provision exposed commissioners’ resource dependence on VCSEs, which led to more collaborative, flexible commissioning. Tracer group characteristics: tracer groups’ characteristics affected commissioning partly through the above contexts; that is, whether VCSEs were few and large (e.g. hospices) or the opposite (social prescribing); undertaking statutorily mandated activity (e.g. hospices, some learning disability support). The degree of specialisation of VCSE work affected the number and mix of providers. Stability of demand for a VCSE’s activity favoured more collaborative commissioning. A temporal context: the COVID-19 pandemic shifted the commissioning of VCSEs substantially towards a more collaborative mode, and this change had not fully reverted. The outcomes of these mechanisms included some that approximated to commissioners’ and VCSEs’ respective aims in using the above commissioning mechanisms. For commissioners the main outcomes were to obtain the use of VCSE resources, sometimes at below the full cost of provision. It was often claimed that VCSE activities had relieved, or would relieve, pressure on NHS services, but firm evidence was scarce. Commissioning VCSEs also offered a way for commissioners to pilot test possible innovations (e.g. for self-help in maintaining health) and a ready-made route to access patient, carer and public opinions. For VCSEs the main outcome was income, although the income flow was often unstable, which made it hard to retain paid staff, and below the cost of their activities. Some VCSEs also increased their voice in the local health system, but we also found a large periphery of VCSEs that did not participate in the networks described above and were not commissioned. We found a more nuanced picture than previous studies’ account of the internal ‘degeneration’ of VCSEs as a result of being commissioned. Most VCSEs reported little change to their aims or ‘mission’, and some had increased the professionalism of their management. We also found emergent, unforeseen outcomes. There was a general shift towards collaborative commissioning away from commodified commissioning, but it made the discursive gap between commissioners and VCSEs more apparent. The undercosting of some VCSE activities amounted to a hidden subsidy to commissioners from VCSEs. VCSE activities added a new category of public health activities, those of preventive self-care at a personal and family level, to the longer-established models of clinical prevention (vaccination, etc.) and the ‘new’ public health (legislative and policy changes, e.g. food labelling controls). ICS formation tended to disrupt co-commissioning networks in the short term, but offered a longer-term prospect of greater VCSE input into co-commissioning. Conclusions By research question: RQ1. Commissioners used two commissioning mechanisms in parallel. Each embodied a different kind of governance structure and a different mode of commissioning. Centred on the commissioning cycle, the quasi-market mechanisms implemented a heavily commodified mode of commissioning whose paradigm was a principal–agent relationship between commissioner and VCSE. The networking mechanisms of collaborative commissioning consisted of a triple negotiated order: networking among commissioners, networking among VCSEs and networking between commissioners and VCSEs. A barrier to using this hybrid mechanism was that at certain points in the commissioning cycle, especially provider selection, the mechanisms conflicted. Unfavourable contexts were other barriers: more specialised VCSE activities, dependence on a few individuals, non-coterminous commissioners, possible competition from corporate and public providers, population dispersal and fiscal austerity. RQ2. For VCSEs, consequences of being commissioned depended on the mode of commissioning but were mainly some extension of VCSE activities, including advocacy; marginal adjustments (not drift) in mission; and additional funding, but often unstable and below the full cost of these activities. RQ3. VCSE involvement in co-commissioning occurred through networking rather than quasi-market mechanisms. VCSEs with long-established working relationships with commissioners were often the ones involved, and were often larger VCSEs. Small local VCSEs were involved vicariously through their participation in VCSE infrastructure bodies or networks which were more directly involved, but a large periphery of small VCSEs remained outside these networks. RQ4. The ACAPs that commissioners needed were a combination of managerial and communication skills: to routinise ‘acquiring’ and ‘assimilating’ information from VCSEs; to specify calls for practical proposals or bids from VCSEs so as to elicit responses from suitable VCSEs; and then to apply (‘exploit’) information and ideas from VCSEs in ways that did not then penalise the VCSEs which provided them (e.g. by excluding them from tendering). VCSEs needed to develop the capacities to supply evidence, both hard and soft intelligence, about why they wished to supplement or amend commissioners’ assumptions and proposals, and to express what outcomes their proposals would offer, what metrics could be used to verify that, and more generally how the VCSE’s proposals would add social value. In ACAP terms, the commissioners appeared to need to develop their knowledge acquisition and exploitation capacities especially, while VCSEs needed to develop their knowledge acquisition and transformation capacities. Implications for health and care Below-cost funding for VCSE activity is not sustainable in the long term because VCSE activities require paid-for inputs, even when volunteer labour is used. The instability of VCSE income from commissioners contributed to VCSE staff turnover. If it continues, commissioners’ move towards longer contracts implies that a stratification and concentration of the VCSE side of the quasi-market may occur. Research recommendations are, in descending priority order, for research to: Quantify the impacts of VCSE activity on demand on NHS services. Extend our analysis of NHS commissioner spending on health-related VCSE activity to include local authority spending, and longitudinally. Test whether our findings apply to the commissioning of large (cross-England) VCSEs engaged in health-related activity. Assess the nature, effects and development of ACAP in commissioners and VCSEs by means that combine representative sample surveys (e.g. using the survey instrument developed for this study) and deeper analysis of the specific mechanisms by which ACAP affects commissioning practice. Funding This award was funded by the National Institute for Health and Care Research (NIHR) Health and Social Care Delivery Research programme (NIHR award ref: NIHR128107) and is published in full in Health and Social Care Delivery Research; Vol. 12, No. 39. See the NIHR Funding and Awards website for further award information.
Keywords
