Historical perspectives

Much historical analysis of the welfare of older people from the 1970s and 1980s explicitly adopted a political economy framework. This is a Marxian-inflected variant of what in the comparative historical literature is termed a structure-functionalist approach, treating the development of social welfare as a concomitant of economic modernisation (Skocpol and Amenta Reference Skocpol and Amenta1986). Welfare states serve essential interests of market societies by reproducing and maintaining human capital and legitimising an unequal social order: ‘a built-in economic and political stabilizer’ (Gough Reference Gough1979; Offe Reference Offe1984: 148). Thus benefits and support for older people serve principally to control the labour market and disguise structural unemployment, with retirement perpetuating existing inequalities; poverty and low status are exacerbated through inadequate resourcing (Phillipson Reference Phillipson1982; Walker Reference Walker1981). In this context institutional responses tend to ‘regulate and confirm inequality’ (Townsend Reference Townsend1981: 22). Moreover, in periods of economic downturn those unproductive groups reliant on state support are most vulnerable to austerity measures (Phillipson Reference Phillipson1982: 2–5). The research agenda has therefore aimed to trace policies which reflected these low priorities and establish the assumptions behind them (Means and Smith Reference Means and Smith1998 [1985]: 8–9).

The ‘division’ of British medicine has provided another structural logic which explains why the health care of older people was relatively neglected. Since the rise of hospital medicine in the early nineteenth century, status and resources increasingly accrued to curative care for acute illnesses, to the detriment of the ‘chronic’ patient (Peroni Reference Peroni1981; Starr Reference Starr1982: 157–9, 171). The concept of ‘chronicity’ was polysemic, variously indicating factors like insidious onset, episodic course, long duration and incapacity from employment, but in Britain it was synonymous largely with age (Peroni Reference Peroni1981: 43–4; Weisz Reference Weisz2011: 442). An acute/chronic distinction was perpetuated by organisational arrangements, with the charity-funded voluntary hospitals refusing to admit ‘incurable’ or potential long-stay cases, which by necessity then fell on public provision – the Poor Law (Abel-Smith Reference Abel-Smith1964: 36–45). The case then is that an ingrained prejudice against geriatric work persisted into the period of the NHS, with the majority's ‘apathy’ and ‘hostility’ confounding efforts by progressive doctors (Bosanquet Reference Bosanquet1978: 126; Honigsbaum Reference Honigsbaum1979: 309; Lewis Reference Lewis2001: 346; Phillipson Reference Phillipson1982: 86–9, 116–9). The impact of geriatric medicine from the national perspective has been assessed within this context, though different positions are taken. Bosanquet argued that the Ministry of Health (later Department of Health and Social Security) consistently supported the development of geriatrics but was thwarted and ignored by medical professionals (1978: 126–42). Bridgen (Reference Bridgen2001), by contrast, accused central government of failing to supply strong leadership and promulgating an unrealistically low ratio of geriatric beds to population, thus allowing regional administrators to neglect development plans. Martin (Reference Martin1995) identified inherent uncertainty over whether geriatrics was a discrete specialty or essentially a social medicine, and she characterised this ambiguity as a discursive manifestation of conflict between professional groups for control of space and resources. There is one regional case study, of 1950s Birmingham, which demonstrates active local leadership surveying chronic sick hospitals and encouraging a medical response, but lacks empirical detail on the outcome (Denham Reference Denham2006).

Another useful approach is the concept of path dependence within the analytical perspective of historical institutionalism. This project, of bringing the state back into the history of social welfare, emphasises the role of governing institutions – the polity, bureaucracy, formal and informal rules, administrative and legal structures – as autonomous forces of change (Skocpol Reference Skocpol, Evans, Rueschemeyer and Skocpol1985). It points to bureaucrats as independent actors, to the ways in which a nation's political system and legislative processes determine the scope for reforms, and to the importance of early choices in establishing ‘policy feedbacks’ which determine subsequent options (Hacker Reference Hacker1998; Immergut Reference Immergut1992). The related idea of path dependence entered policy studies from the economics of innovation, and the problem of explaining how arguably sub-optimal products (like the QWERTY keyboard or the VHS video system) might still dominate markets, in contrast to the expectations of classical economic theory. This is due to mechanisms of ‘increasing returns’, such as the high fixed costs of initial investment, and the adaptation, and hence expectations, of consumers, which militate towards continuity (Arthur Reference Arthur1989). In political science the metaphor implies the constraint on future choices by the early adoption of a particular policy trajectory (Pierson Reference Pierson2004). The feedback effects of these early choices may also include ‘non-increasing returns’, such as the costs of early investment which lock governments into a course of action even when appealing alternatives emerge, or the creation of interest groups who reshape the policy arena, or of administrative structures which establish or restrict dynamics of change (Kay Reference Kay2005). In essence the method entails identification of a critical juncture at which a trajectory of development is launched which becomes self-reinforcing: henceforth ‘the costs of switching to some previously plausible alternative rise’ (Pierson Reference Pierson2004: 21).

Despite criticism of its theoretical underpinnings, and the reservation that it explains continuity more readily than change, path dependence has been fruitfully applied in histories of health policy. For example, in Britain initial investment costs and consumer adaptation to a tax-funded NHS can explain the repeated defeats of proposals for change to a social insurance model; in the United States interest-group feedback effects help clarify why the early rejection of mandatory health insurance made subsequent reform efforts unlikely to succeed (Hacker Reference Hacker1998; Lowe Reference Lowe, Gorsky and Sheard2006). In relation to older people, Bridgen and Lewis tentatively evoke path dependence in discussing the political and administrative tensions of the ‘health/social care boundary’ since 1946 (1999: 15–16). Their themes are the recurrent issues of ‘bed-blocking’ and cost-shunting which arose because responsibility for the medical and social needs was divided between the NHS and local authorities. The implication, which they do not develop, is that this damaging arrangement became locked in by decisions at the start of the NHS.

This literature therefore suggests three subsidiary problems through which my main research question may be explored. First, does the regional experience confirm the orientation of welfare policies in the field towards productivism and neglect of older people outside the labour market? Second, what place was accorded to geriatric medicine within the structure and resource priorities of the NHS, and what factors determined this? Third, were efforts to implement a national reform agenda confounded by the policy feedbacks of earlier decisions?

The region and the pre-NHS inheritance

The case study area is a part of the South Western RHB, one of the NHS administrative units overseeing the hospital sector in the period between inception in 1948 and reorganisation in 1974. It incorporates two counties, Gloucestershire and Somerset, and three cities, Bristol, Bath and Gloucester. Bristol was the region's capital, a port, transport hub and industrial centre, home to a university, medical school and a cluster of general and special hospitals, including a teaching hospital. Bath's past as spa and leisure town had also bequeathed extensive hospital provision originating in the philanthropic voluntary sector, while Gloucester was a smaller port city with fewer facilities. The counties were largely agricultural, though with a (nearly exhausted) coal industry and limited tourism; many rural residents still remained fairly remote from urban services. Population grew from about 1.27 million in 1931 to 1.76 in 1971, while the numbers aged over 65 rose from about 120,000 to just over 250,000, a growth of from 9 to 14 per cent of the total (Census of England and Wales 1931, 1971).

The institutions which became the RHB's geriatric hospitals or accommodated its ‘chronic’ beds had mostly originated within the Poor Law. There had once been 35 workhouses in the region, of which the majority (29) were built between 1836 and 1847, following the 1834 Poor Law Amendment Act. A plentiful visual record remains (available electronically) of most of these, partly confirming familiar Dickensian tropes, with barrack- or prison-like facades and panopticon or cruciform designs advertising regimes of deterrence; a few though appear more akin to charity almshouses or country houses (Higginbotham Reference Higginbotham2011). The original intent of instilling work discipline into the feckless poor was giving way to the long-term accommodation of older or mentally ill people by 1900: a return of 1904 recorded 60 per cent aged over 60, of whom the majority were ‘infirm’ (Parliamentary Papers 1904: 34). The 1930s saw the closure of most ‘casual’ wards for tramping workers or the homeless, and qualitative evidence, for example from Somerset, reveals that despite the nominal distinction between ‘house’ and ‘infirmary’ wings of the institutions, most inmates by the 1930s had ‘some physical or mental infirmity’ (Somerset County Council 1938). By 1948, 27 of the institutions were still extant, of which 22 became NHS hospitals, three old people's homes and two ‘joint use’, shared by the NHS and local authorities; 12 of these hospitals were geriatric, and three general, including chronic beds.

How far did the Local Government Act (LGA) of 1929 deliver a more integrated system in which health care shed the Poor Law taint? The policy goal was for councils to transfer hospital responsibilities away from public assistance (as the Poor Law was renamed) to the control of its Public Health Committee, under medical stewardship and integrated with other services overseen by the Medical Officer of Health (MOH). In the event only five institutions were thus ‘appropriated’, with only Bristol and Gloucester creating municipal general hospitals. In Bristol this was Southmead Hospital, which also had a large chronic bed complement, alongside two public assistance institutions (PAIs), Eastville and Stapleton, the latter particularly accommodating older infirm people. In the counties the majority of the mixed workhouses continued as before.

The record of local government stewardship prior to the NHS affirms the proposition that economically unproductive groups lost out in the competition for health resources. Table 1 compares health spending from local taxation in the peacetime decade following the LGA, aggregating the relevant departmental budgets of the five councils. ‘Health’ is defined to include only personal, rather than environmental, interventions. During these austerity years the Public Health Committees, with their focus on mothers, children, acute hospital care and tuberculosis, saw real expenditure rise as did the school medical service, introduced in the pre-war Liberal welfare reforms; taken together (Table 1, column c) real spending rose from £0.37 per capita to £0.53 by 1939. Public assistance expenditure on institutions initially dipped, reflecting appropriation-related budgetary transfers, then began to rise modestly; however, the growth was mainly driven by progressive Bristol; if the city is excluded (Table 1, column e) it is clear that the proto-geriatric hospitals/residential homes saw declining spending until 1937. Qualitative data from wartime survey reports duly testify to poor fabric and facilities, and inadequate medical care (Cope et al. Reference Cope, Gill, Griffiths and Kelly1945: 131–2).

Table 1. Personal health expenditure (£) per head of population, South West region, constant prices (1929)

Notes: PHC: Public Health Committee: maternity and child welfare, tuberculosis service; isolation/general hospitals; mental hospitals (proportion); mental deficiency; welfare of blind; vaccination; venereal disease services; administration. SMS: School Medical Service; inspection, treatment, institutions. PAC: Public Assistance Committee: mental/general hospitals (proportion); institutions.

Sources: Abstract of Accounts Gloucestershire, Somerset County Council; Bristol, Bath, Gloucester County Borough Council, 1930–31 to 1939–40.

Institutional factors and path dependence help explain this. The British polity had long devolved social welfare powers to local government within a permissive framework, where, in the name of local autonomy, latitude was extended to councils over decisions about appropriation, finance and control. Central encouragement had been channelled instead through a grant system which in the 1920s had bolstered services specifically for mothers, children and tuberculosis sufferers. However, localism persisted after 1929 in Poor Law policy. A concern to honour the voluntary efforts of ex-guardians persuaded legislators to retain unelected, co-opted members on the new Public Assistance Committees. This meant that many personnel simply continued in place: a feedback effect traceable to the 1834 Poor Law legislation, coupled with the permissive tradition. In Bath, for example, there were 17 ex-guardians on the new 25-person committee in 1929. In this environment it is unsurprising that officials were (in the eyes of Ministry observers) often ‘deep-rooted in Poor Law ways’, or of a ‘carry-on character’ and unreceptive to change (Parsons Reference Parsons1931–2: 127, 189–90, 200; 1932: 127).

With development thus stalled, the war broke out. Its impact, as Means and Smith have shown for London, was to exacerbate the problems faced by older welfare recipients (1998 [1985]: 13–59). In the South West the long-stay sector faced a critical shortage of nursing staff and the co-option of beds for military or civil emergency purposes. In Bristol, for example, Southmead accommodated ‘chronic’ and some ‘sub-acute’ evacuees from London, while a recently opened old people's home was converted into a hospital for long-stay and venereal diease patients. Meanwhile the nursing shortage at Stapleton became increasingly severe and calls to alleviate this by expanding domiciliary nursing led nowhere, as did appeals to the Ministry of Labour to fill vacancies for hospital orderlies and domestic staff (Bristol Medical Officer of Health (BMOH), 1943, 1945). In Gloucester, the older sick were moved to hutted accommodation, and in rural Gloucestershire were relocated to institutions far from their homes and social networks (Cope et al. Reference Cope, Gill, Griffiths and Kelly1945: 12, 21). Here the scarcity of domiciliary nurses pushed older people into institutions, where by 1945 staff shortages became so drastic that Gloucestershire Public Assistance Committee called for full indemnity of its officers in case of their failure to ‘carry on a public service’ (GPAC 1940–48: February, March 1945). Against this bleak picture there was only one positive gesture towards the universalist social policy promised by the 1942 Beveridge Report. This was the symbolic renaming of institutions, so that for instance in Gloucestershire the ‘Public Assistance Institutions’ became ‘County Infirmaries’, excepting Tewkesbury's, which was rebranded as ‘Holmwood’ (GPAC 1940–48: March, May, June 1945).

NHS structures

The creation of the NHS is widely perceived as a fundamental rupture with the past. In the hospital sector its radicalism lay in the ‘nationalisation’ of the voluntary hospitals and the removal of the public hospitals from local authority control. All would now be funded principally from general taxation and organised in a hierarchical administrative structure accountable to the Minister of Health. The assumption was that regionalisation would facilitate unified planning, developing services in a coherent manner to address local needs (Webster Reference Webster1988: 262–5). At the lower level a tier of Hospital Management Committees (HMCs) was set up, many overseeing a group of institutions. Above the HMCs were the RHBs of which the South Western covered Somerset, Devon, Cornwall and parts of Gloucestershire and Wiltshire. Transfer of institutions to the NHS appears to have been a consensual process in the region, with the ex-PAI hospitals assigned to HMCs where they were integrated under single authorities. For example, the South Somerset HMC brought the Yeovil and Wincanton institutions alongside one general hospital, a maternity home, an isolation hospital and two cottage hospitals (South Western Regional Hospital Board (SW RHB) 9/1/1948). The exception was Bristol's Stapleton, which became the region's major geriatric hospital with its own HMC (SW RHB MP 1949).

In other respects though, the NHS settlement maintained and enforced pre-existing divisions, thanks to the realpolitik of the Act. Regionalisation had emerged not as the preferred option of planners but rather as a compromise which satisfied interested parties (Klein Reference Klein2006 [1983]: chap. 1). The alternative model, with support in the Ministry and the Labour Party, was to build the service on local government, which would have had the advantage of truly integrating the hospital service with the public health and social care work already undertaken by councils. However, the British Medical Association (BMA) was resolutely opposed to this, fearing doctors would become salaried officials working at the behest of local bureaucrats. Bevan's solution was a tri-partite structure in which the administration of hospitals was distinct from that of primary care and also from local authorities which maintained only limited public health functions. Meanwhile the National Assistance Act (1948) terminated the Poor Law, abolished the term PAI, and obliged local authorities to develop services for older people in need, with Part III of the Act specifying residential accommodation (Means and Smith Reference Means and Smith1998 [1985]: chap. 4). Thus not only was the administrative separation of social care from the hospital sector perpetuated, but now the two were divorced in budgetary terms.

Path dependence conditioned by already dominant interest groups can also be discerned within the RHB itself, where the pre-1948 voluntary hospital leadership was a significant force. This was because of Bevan's concern to placate opponents of hospital nationalisation by avoiding the appearance of political bias. This led to the marginalisation of Labour figures hitherto active in municipal health politics and of members of the organised working class who had participated in inter-war hospital management as trade union or contributory scheme representatives (Gorsky and Mohan with Willis Reference Gorsky, Mohan and Willis2006; Webster Reference Webster1988: 274–82). Instead, establishment figures already prominent in hospital governance, academia or acute medicine loomed large. At the South Western RHB's first meeting, the chair, a Bristol industrialist, assured the board their work would be an ‘extension’ of the ‘voluntary system’, which would preserve its ‘great traditions’, and that it would have ‘almost complete freedom of action’ (SW RHB 18/7/1947). Indeed much official reassurance was given about local autonomy within a devolved system (Bristol and District Divisional Hospitals Council 1947; Webster Reference Webster1988: 95).

Potentially then, the NHS structures augured continuity with earlier arrangements that had privileged acute over ‘chronic’ medicine and entrusted much responsibility to the Poor Law. Nonetheless, local hospital planners voiced early commitment to improving the regional distribution of services and addressing the poor fabric of the ex-PAIs (SW RHB 18/7/1947, 29/4/1949). Similarly, Bristol University and the city's pre-NHS Hospital Council convened a conference in 1947 on the ‘Elderly and Infirm’ and began lobbying for specialty geriatric care and better domiciliary services (Bristol and District Divisional Hospitals Council 1948). Improvement therefore seemed imminent.

The promise of geriatric medicine

The early years of the NHS saw an upsurge of interest from doctors in the health and treatment of older people. The 1947 Bristol conference was triggered by an influential BMA report, proposing to end the workhouse ethos of ‘therapeutic nihilism’; instead regional planning schemes should allocate patients either to a centre which could provide intensive medical or nursing care, or to long-stay annexes for palliative care. (Means and Smith Reference Means and Smith1998 [1985]: 119–21). Arguably this sudden fascination with geriatric medicine in part reflected the self-interest of consultants concerned to rationalise the distribution of long-stay patients and clearly delineate NHS responsibility for older people from that of local authorities (Martin Reference Martin1995). However, it also arose from the clinical work of pioneers such as Marjory Warren in London, who had been brought into contact with older patients following the municipalisation of large PAIs (Jeffreys Reference Jeffreys, Bornat, Perks, Thompson and Walmsley2000).

Locally a new optimism was much in evidence. Geriatrics arrived in regional medical discourse with a spate of articles in the South West's leading professional journal, as well as publications by local doctors in national outlets. A signal indicator of rising interest was the invitation to eminent geriatrician J. H. Sheldon to give the prestigious Long Fox lecture at Bristol Medical School in 1957 (Sheldon Reference Sheldon1958; cf. Sheldon Reference Sheldon1948). Such expositions typically began by discussing the changing demographic context, a reflection of the intellectual influence of the 1949 Royal Commission on Population (Andrews Reference Andrews1951; Hughes and Pugmire Reference Hughes and Pugmire1952). Writers then castigated the failings of the present system, arguing that under-diagnosis was rife, that general practitioners (GPs) were poorly trained for this aspect of medicine, and that many bedridden patients had become so not from irremediable causes but from very poor nurse/patient ratios and consequent neglect (Andrews Reference Andrews1951: 37; Stewart-Smith Reference Stewart-Smith1956; Wilson Reference Wilson1951; Wofinden Reference Wofinden1952: 564). Next came a call to arms, championing effective new treatments, such as bladder control techniques for incontinence and suspension therapies for bedbound arthritic and hemiplegic patients, and urging take-up of the free provision of NHS spectacles, hearing aids and chiropody services to facilitate independent living (Sheldon Reference Sheldon1958: 74–5; Wilson Reference Wilson1951: 38–42). The prescriptions became less certain when it came to institutional care. Ideally, the ‘grim, grey buildings’ of the PAIs should be superseded by public provision of new homes, hostels, hospitals and domiciliary services (Andrews Reference Andrews1951: 36). But given the division of responsibility between RHB, the HMCs, and both the health and welfare departments of local government, how were older people to be allocated to appropriate accommodation? All agreed this was highly problematic, as ‘one class shades imperceptibly into the other and … there is no definite borderline between those problems which are social and those which are medical’ (Wofinden Reference Wofinden1952: 567–8; also Anon. 1955: 20). Nonetheless these articles reflect a new confidence that the problems were now identified and articulated, and the NHS represented a fresh start.

A local take-off of geriatric medicine seems to have hinged on the presence of key individuals (Denham Reference Denham2006; Thane Reference Thane2002: 449). The South West illustrates this, in the person of William Hughes, the area geriatrician based at Bristol's Stapleton Hospital. He first introduced a system of classification, delineating ‘mental defectives’ (progressively moved to mental hospitals), ‘psychotics’ (mental patients certified under the Lunacy Acts), ‘ambulants’ (patients with ‘a mild degree of mental deterioration’ who required hostel or hospital accommodation), and ‘frail ambulants’ (patients with some ‘mental or physical infirmity’ needing care by skilled staff and potentially suitable for transfer to hostels) (Hughes and Pugmire Reference Hughes and Pugmire1952: 1249–50). At the same time he established a geriatric centre, including beds for ‘active treatment’ and a physiotherapy department, and increased medical staff from two to six, alongside a full-time almoner to liaise with local government, voluntary sector and families (Hughes Reference Hughes1951; Hughes and Pugmire Reference Hughes and Pugmire1952: 1251). Hughes' vision was that Stapleton would function as the region's geriatric centre, while the smaller rural ex-PAIs would become its peripheral hospitals, both housing long-stay patients and referring ‘heavy nursing’ cases back to Bristol (Hughes and Pugmire Reference Hughes and Pugmire1952: 1252).

Surviving activity data published by Hughes (Table 2) suggest that his policies benefited patients, on the assumption that the volume and pace of discharges indicates an active treatment regime (cf. Bridgen Reference Bridgen2001: 519–20). About 10 per cent of admissions were treated and discharged within a month (7 per cent in the first two weeks). A further one in three was discharged by eight months, pointing to some successes of more active therapy and to the impact of antibiotics. This suggests that the ubiquitous usage before 1948 of the label ‘chronic sick’ hospital is problematic, and that the chronic/acute dichotomy was in part a social construction born of inadequate medical provision. However, an even greater proportion died in the hospital, and thus another key role was the management of terminal illness; one in ten admissions was dead in the first two weeks, but many lingered and this suggests a hospice function, providing palliative social care. Causes of death were not reported, though post mortem data from Exeter in the same period are suggestive: here principal causes of mortality in the over-sixty-fives were coronary heart disease, arteriosclerosis and hypertension, neoplasms, prostate and kidney disease, abdominal crises (hernia, perforated ulcer, etc.) and traumas (probably from accidents in the home) (Stewart Smith Reference Stewart-Smith1956: 46–7). Stapleton also acted as a clearing house for mentally ill patients, who were mostly certified within the first week and passed to other institutions. About 12 per cent of admissions remained after eight months: this was the core of ‘long-stay’ patients.

Table 2. Length of patient stay, admissions to Stapleton Hospital, June 1950 to May 1951

Source: Hughes and Pugmire (Reference Hughes and Pugmire1952).

These early initiatives informed broader planning for geriatrics. In the early 1950s the RHB produced a series of plans for the major specialties (such as tuberculosis, maternity, ophthalmics, dermatology, ear, nose and throat, orthopaedics, paediatrics, etc.) which proposed locations of regional and subsidiary centres of expertise, along with requisite bed numbers and staff appointments needed to ensure coverage. The Board duly appointed a geriatrics sub-committee in 1951, chaired by Bristol's Alderman Milton, a veteran of the appropriation of Southmead (SW RHB 2/6/1950, 13/7/1951). Its report was a blueprint for an integrated medical service for older people in the Bristol clinical area (South Western Regional Hospital Board Geriatric Sub-Committee 1951). This endorsed Hughes' approach, delineating a central geriatric unit for active treatment, rehabilitation and support of domiciliary care by GPs and local authorities, as well as peripheral hospitals specialising in long-stay care for either mental or physical illness. The emphasis on Bristol as regional centre built upon inter-war developments, when the city's diverse Poor Law inheritance had permitted rationalisation of functions and patient types across the three municipal institutions, creating a large proto-geriatric hospital unlike the PAIs in the smaller cities and towns. However, it also reflected the principle of hierarchical regionalism in shaping the organisation of the other specialties noted above. With respect to broader principles, the plan affirmed Sheldon's argument that preventive and domiciliary care should be the frontline of geriatric medicine, with the hospital in an ancillary role (Sheldon Reference Sheldon1958: 73–4; Wofinden Reference Wofinden1952: 563). It was also the first clear manifestation of the RHB's concern with the contested nature of health/social care responsibilities under the divided tripartite system, centring on the indeterminate status of the ‘frail ambulant’.

The Bristol area plan laid the foundation for a policy document for the whole region (SW RHB 4/7/1952). This acknowledged that the built legacy of the Poor Law was a major impediment to change and fully accepted contemporary arguments about the potential of active therapy to remedy failures of inter-war care. It then mapped out the solutions: better patient classification to unscramble the mixing in ex-PAIs; geriatric units affiliated to major general hospitals, appointment of consultant geriatricians in each clinical area, new long-stay accommodation and joint RHB/local authority liaison committees. Again though, the document frankly expressed the rising tensions with local government over what the RHB claimed was ‘bed-blocking’ and the ambiguous status of ‘frail ambulants’: ‘large numbers of relatively active patients’ were occupying beds ‘which are needed for the admission of urgent cases’ due to ‘failure to provide suitable accommodation elsewhere…’.

In sum, the early years of the NHS saw regional administrators addressing the programmes advocated by the pioneers of geriatric medicine. But even as they did so a series of obstacles emerged. The tripartite system militated against rational planning, and the temptation to shunt costs lent a political subtext to what should have been clinical judgements (Wofinden Reference Wofinden1952: 567–8). The crux of the matter, however, was that the expansion and renovation of hospital and residential accommodation hinged on the possibilities for capital investment.

The limits of reform: expenditure and planning

Viewing developments from a national perspective, commentators have characterised the next two decades as a period of policy drift. Geriatric medicine remained a marginal specialty whose interests were easily overruled on regional boards (Bridgen Reference Bridgen2001: 509, 511, 520; Martin Reference Martin1995: 452–9). The Ministry of Health failed to give a strong lead, despite various enquiries into older people's health needs. Instead its policies constrained hospital improvement by insisting that new beds were unnecessary, but without providing the resources to build up domiciliary care or local authority accommodation (Bridgen Reference Bridgen2001: 513–20). Indeed, alternative policy options to institutionalisation foundered on official scepticism about the value of funding such community services (Thane Reference Thane2002: 453). The national context for inaction was a lack of investment in health expenditure in the 1950s and beyond. As Webster shows, both current and capital expenditure fell in real terms, remaining below their 1950 levels until 1957, from when they gradually increased, doubling by 1973. Even when expanding, the NHS suffered in comparison to other areas of social spending; its share of UK public expenditure fell in the early 1950s, then stayed roughly static at between 19 and 21 per cent, in contrast to the gains made by education and social security. And, as a percentage of gross domestic product, its budget only regained its 1949–50 level by 1964–65 (Webster Reference Webster1996). What does the regional evidence reveal of the impact of this parsimony?

The South West had inherited almost twice as many ex-PAIs as other RHBs and, as noted, many had been poorly maintained by local authorities. There was some limited capital investment early on, for example in Stapleton's geriatric unit and in Chipping Sodbury's ex-PAI, but while not entirely neglected these hospitals were not a development priority (SW RHB 2/6/1950). By the 1960s, for example, nine ex-PAIs were still without lifts and non-ambulant patients had to be carried up and down stairs (SW RHB 1/2/1963).

With respect to current expenditure, the RHB based its allocations largely on patterns established before the NHS. Analysis of trends in funding geriatric hospitals is difficult because such data are only available at the level of HMC. Only the Stapleton HMC was exclusively long-stay/geriatric, and its experience indicates a declining commitment. In 1951–52 its percentage share of the region's expenditure on hospital staff stood at 3.57, and this declined to 2.97 in 1963–64 then 2.66 by 1967–68; its share of total HMC expenditure similarly dwindled over the same dates, from 2.97, to 2.83, to 2.67 (South Western Regional Hospital Board Finance Committee (SW RHB FC): passim). The enduring differential was also evident in a comparative survey of hospital food from 1956, which revealed the long-stay hospitals' cheaper dietaries, deficient in calorific content and heavy on dull, bland foods (SW RHB FC 21/9/1956). Although the Board blamed central under-funding, it was also deaf to internal calls for reallocation of resources. In 1954, for example, its Mental Health Committee asked that mental hospitals, which contained over half of the region's beds, should be allocated 40 per cent of capital expenditure: hitherto they typically received less than half that. No action was taken, after medical representatives on the RHB argued that this was ‘fantastic’ and, if implemented, the cut to the acute sector would drastically demoralise the nursing staff (SW RHB 15/1/1954).

Thus the management board influenced by the old voluntary hospitals leadership and by the interests of acute medicine tended to hold pre-NHS resource allocations constant. Against this status quo there is sporadic evidence of a geriatrics lobby working through sub-committees, but the very repetitiousness of their reports indicates how little was changing. In 1958 there were calls for more accommodation, particularly for the mentally ill, more registrars with geriatrics training, and better co-ordination between GPs and ancillary services to strengthen domiciliary care (SW RHB 10/10/1958). Five years on, the same refrain was heard, when the psycho-geriatric committee criticised the persistence of mixing and misplacement of mentally ill patients (SW RHB 11/10/1963). As before, the ideal was better facilities for assessments, new long-stay accommodation and better joint working between the responsible parties. However, the backdrop to this was the inadequacy of local authority services. As with the NHS, councils wishing to develop residential services had faced capital constraints in the 1950s, and subsequently were reluctant to build capacity for fear this would simply encourage early discharge from the NHS (Lewis, Reference Lewis2001: 347; Means and Smith Reference Means and Smith1998 [1985]: chap. 5). Community care similarly evolved at a slow pace, with the domiciliary nursing and home help service catering also to mothers, children and the acute sick. In the three South West cities real spending on these two areas doubled between 1950–51 and 1968–99 [£186.6 to £363.63 per 1,000 persons (1948 prices)], though that on the health visiting service, which might have taken the lead in assessing needs and contacting appropriate agencies, remained flat (£60.28 to £73.18) (calculated from Institute of Municipal Treasurers and Accountants (Incorporated), The Society of County Treasurers 1951, 1970; cf. Means and Smith Reference Means and Smith1998 [1985]: 285–8).

Regional geriatrics champions therefore proved unable to shift the RHB's policy. Geriatric medicine also remained of low profile in primary care, with, for example, GPs failing to refer old people for glaucoma and operable cataracts because they assumed age was a bar to surgery (SW RHB 8/10/1954). The ex-PAIs were unable to develop new identities, so that in Somerset for instance, Wells Infirmary continued to combine ‘mental deficiency’ patients with the physically sick in a manner reminiscent of Poor Law mixing. Others, such as Clutton, mixed residents and patients just as in the workhouse era, and continued to be jointly run with Somerset County Council. Here the ‘bed-blocking’ stand-off was enacted within a single institution as the local authority pleaded with the HMC for separate accommodation for residents with senile dementia, and the HMC pleaded with the Ministry for more resources for domiciliary care (North East Somerset Hospital Management Committee 29/4/1950, 26/3/1954, 25/6/1954). Similarly, the continuing presence of ‘chronic’ sick in acute wards at Bath's St Martin's led to calls for the city council to expand hostel provision, driven partly by the assumption that only acute care would ‘maintain the interest of the nursing staff’ (Bath Hospital Management Committee 28/1/1949, 30/6/1953, 28/5/1954). In 1961, the regional development plan considered bed numbers in South Somerset to be satisfactory but found ‘little regard to the rehabilitation of patients’ due to lack of specialists (SW RHB 29/5/1961). In the Bath clinical area a single geriatrics consultant covered the city and parts of North Somerset and Wiltshire, and here the comparatively high ratio of geriatric beds to population concealed poor turnover rates due to under-staffing (Anon. 1970).

The degree of institutional stasis which national and local policy had engendered by the 1960s is illustrated by the limited ambition of the Hospital Plan. Produced in 1962 this marked the start of a period in which technocratic planning was finally given its head and the austerity of the 1950s gave way to a sustained capital programme (Klein Reference Klein2006 [1983]: 47–9, 54–61). The first step entailed local reappraisal of the region's institutional accommodation leading to proposals for renovation (Mohan Reference Mohan2002). The RHB now proposed moving geriatric units into general hospitals, a model which had initially been championed by Warren, and proposed in the Birmingham region; this also mirrored trends in mental health, where the Plan inaugurated asylum closures and new psychiatric units in district general hospitals (McKeown and Lowe Reference McKeown and Lowe1952; Warren Reference Warren1943). In Bristol, the long-term goal was to replace Stapleton (now renamed Manor Park) with wards at Southmead, Frenchay and a proposed new South Bristol Hospital (SW RHB 29/5/1961). For Gloucester, the plan was to close three existing hospitals and start a geriatric unit at the Gloucester Royal, while in South Somerset a unit would open at the Taunton and Somerset to replace closures of smaller establishments (NHS 1962: 185–204).

The limited changes which the Hospital Plan actually signalled are illustrated by Table 3 which analyses geriatric bed distribution across different types of hospital, according to their usage prior to the NHS. It includes the actual beds in 1960 and the projected beds under the Plan, in the North Gloucestershire, Bristol, Bath and South Somerset clinical areas. After 12 years of the NHS, geriatric care remained in institutions which had been in the public sector before 1948. Three-quarters of the beds in 1960 were in ex-PAIs, despite the long-standing claims about their unsuitability. New capacity had not come from purpose-built accommodation, but from the co-option of 13 ex-isolation hospitals, now redundant following the decline of mortality from infectious diseases. Only one ex-voluntary, in Chard, had a small geriatric unit. Nor did the Plan propose to alter this substantially. Although the establishment of geriatric units at general hospitals would have brought some 10 per cent of beds into this category, almost three out of every five beds would still be in ex-PAIs. So again, the keynote of this period is continuity with the era of the Poor Law.

Table 3. Distribution of geriatric hospital beds in the South West region according to pre-National Health Service usage of institutions

Source: National Health Service (1962).

The limits to reform: the performance of geriatric hospitals

Thus far the argument about early initiative frustrated by financial constraints has been developed largely from qualitative data in administrative records. Can this be supplemented by evidence of hospital performance? Unfortunately systematic recording of activity in the early NHS was rather limited, although it is possible to calculate for each hospital an annual turnover rate (in-patients discharged or died per available bed) and its mean length of stay in days. These reveal nothing about quality of care, but they were understood as a basic measure of productivity (SW RHB FC 24/9/1954). With respect to geriatrics, we may also, following Bridgen, treat a rising turnover rate ‘as suggesting a growth in the influence in hospitals of active treatment techniques’ and improvements in specialist staffing (2001: 519–20). It might also be determined by availability of community care facilities for discharged patients.

Turning first to the major geriatric centre, Bristol's Manor Park (the renamed Stapleton), Figure 1 presents these data, as well as total bed provision, so that the effect of capacity on the other measures may be appreciated. Turnover rose and length of stay fell markedly at the start of the sequence, when Hughes introduced his regime of assessment and active therapies described above, although up to 1951 the trend is also explained by bed reductions and transfers to other agencies (Hughes and Pugmire Reference Hughes and Pugmire1952: 1251). Continuing improvement in turnover and a gradually falling length of stay are also discernible up to about 1957. Again this may be ascribed to the impact of geriatric medicine, as the hospital added special departments, first physiotherapy and chiropody, then occupational therapy (Manor Park Hospital Management Committee 1953) and electro-cardiography (Manor Park Hospital Management Committee 1954).

Figure 1. Provision, length of stay and turnover in Manor Park (Stapleton) geriatric hospital, 1948–76.

However, the late 1950s through to the late 1960s saw Manor Park's turnover falling and length of stay rising while bed numbers remained fairly stable. Thus by this point the new methods had achieved all they could, either in treating patients more effectively or in reallocating some to other institutions. Even then, Hughes' stewardship saw continuing augmentation of specialist care, with the opening of an audiometry (hearing aids) clinic (1962) and departments of radiology (1960), speech therapy (1963) and optical dispensing (1965). Not till the late 1960s was there another sharp rise in turnover and a pronounced fall in duration of stay. This followed Hughes' retirement and the appointment of a new area geriatrician, with three additional consultant geriatricians, one at Manor Park and the others at the leading general hospitals, Southmead and the United Bristol (the teaching hospital). At the same time a psycho-geriatric unit was set up to provide psychiatric assessments more rapidly and speed transfers elsewhere (Joint Advisory Committee 16/11/1970). Change was driven by a joint committee on geriatric services for the Bristol clinical area, in which Bristol MOH Robert Wofinden worked with city councillors and geriatricians to exert pressure on the RHB and the local authority. As in the 1950s though, the committee remained frustrated with the low level of investment, both in geriatric hospital facilities and in local authority residential homes (Joint Advisory Committee 20/5/1970).

In the smaller ‘peripheral’ rural hospitals the picture of stasis in the 1950s and 1960s is broadly confirmed (Figure 2). For these institutions some statistical data extend (for reasons unknown) beyond the reorganisation of the NHS in 1974, allowing a longer-term view. Here annual means have been calculated based on the aggregated experience of seven hospitals. This method produces a very uneven result for length of stay, which could fluctuate significantly from year to year in a small hospital depending on the patient mix, so the series is smoothed with a five-year moving average. Mean bed numbers across these hospitals remained flat throughout the sequence. In contrast to Bristol's Manor Park, turnover did not markedly improve in the 1950s and length of stay rose, probably indicating the absence of consultant geriatricians. Gradual improvement began during the 1960s, a time of greater Treasury support for NHS expenditure, and this probably reflects increasing availability of active treatment in special clinics and departments. For example, in 1959 a geriatrician was appointed to Trinity Hospital (the ex-PAI) in Taunton, and in the period 1957–61 rates of new patients trebled in its physiotherapy and occupational therapy clinics (Taunton Hospital Management Committee 1948–65). This was also a period in which the practice of physiotherapy was evolving away from passive massage techniques to make greater use of remedial gymnastics and to emphasise positive mental attitudes (SW RHB 12/10/1956). The early 1960s also saw two specialists appointed in the South Somerset clinical area, ‘for the assessment and active treatment of geriatric patients’, with a view to speeding the closure of the small ex-PAIs and concentrating geriatric beds (SW RHB 29/5/1961). Despite all this, it was only in the 1970s that far-reaching change occurred, with turnover rates increasing by a factor of three, 1970–83, and length of stay falling from about 300 to about 60 days. This is probably attributable to the gradual, if unspectacular, advance of domiciliary support services and to the boost in capital expenditure on local authority residential homes from the late 1960s, at last easing pressure on NHS beds (Means and Smith Reference Means and Smith1998 [1985]: 205–8).

Note: 5 per Mov. Avg.: five-year moving average.

Figure 2. Provision, length of stay and turnover in selected geriatric hospitals, South Western Regional Hospital Board, 1948–86.