Tuberculosis and ethnicity in England and Wales, 1950 ±70 John Welshman

Transcription

1 Sociology of Health & Illness Vol. 22 No ISSN 0141±9889, pp. 858±882 Tuberculosis and ethnicity in England and Wales, 1950 ±70 John Welshman The Institute for Health Research, Lancaster University Abstract This article seeks to contribute to recent debates about ethnicity and health by exploring the history of migration and tuberculosis in England and Wales between 1950 and It concludes that the story was more complex than recent writing, with its emphasis on `port health' concerns, has implied. The fear that tuberculosis was being imported by migrants was certainly a central concern of both early researchers and the medical establishment. However, some researchers did show some interest in material explanations and in the roles of housing and work patterns in the transmission of the disease. A system of medical examinations at the ports of entry was not in fact implemented and it was at the local level that a system of surveillance was set up. Finally, despite much debate about the susceptibility of migrants, racial concerns were less evident than recent writers have suggested. Keywords: Tuberculosis, ethnicity, `port health', Leicester Introduction Recent work on ethnicity and health has attempted to provide new directions for theory and research. Chris Smaje, for example, has argued that analyses of the ethnic patterning of health have often failed to examine the social meaning of ethnicity, and have too often concentrated on binary oppositions that are ultimately unhelpful. He argues that instead of contrasting material and cultural explanations, or placing sociological approaches against those of epidemiologists, researchers should seek to provide a more nuanced picture, not least through developing new concepts of ethnicity. Research should also focus on culture and identity, and many insights are lost if ethnicity is `simply emptied into class disadvantage' (1996: 153). In a similar vein, James Nazroo has examined three different approaches to ethnic inequalities in health ± ethnicity as untheorised epidemiology, as structure and socio-economic status, and as identity. He argues that ethnicity needs. Published by Blackwell Publishers, 108 Cowley Road, Oxford OX4 1JF, UK and 350 Main Street, Malden MA 02148, USA.

2 Tuberculosis and ethnicity in England and Wales, 1950 ± to be adequately theorised and recognised as both structure and identity. Nazroo suggests that exploring ethnicity should allow a dynamic exploration of culture and the relationship between culture, context and class (Nazroo 1998). This article seeks to contribute to current work by providing an exploration of the ways that notions of race, socio-economic deprivation and culture were invoked in earlier debates about ethnicity and health. The issue of tuberculosis and migration ± in England and Wales, between roughly 1950 and 1970 ± provides a suitable case study. Recent concerns about the spread of drug-resistant tuberculosis have been well documented (Tanne 1999). In this context, some observers have argued that systems of detecting tuberculosis in new arrivals to the United Kingdom need to be improved. Current regulations state that anyone planning to live in the United Kingdom for over six months who arrives from areas defined as high risk (where the tuberculosis incidence is more than 40 cases per 100,000 population per year) should be screened. Consultants in communicable disease control in health authorities in which the migrant is planning to live should be contacted. It is up to them to contact migrants or asylum seekers, carry out follow up tests to find people registering positive on skin testing and those requiring vaccination, and to initiate chest radiography for those who did not have it at port health units (Joint Tuberculosis Committee of the British Thoracic Society 1994). However, many port health units no longer have the resources to deal with migrants and asylum seekers, and most health authorities are unable to offer comprehensive contact tracing and screening. Although in practice it is general practitioners who have to deal with the health concerns of these new arrivals, studies indicate that they are not initiating screening either. Thus it is suggested that both screening and follow up need to be improved (Hargreaves 2000). It is, however, also arguable that earlier explorations of the background to the setting up of screening systems and of the history of ethnicity and health in general, have been limited through their dependence on the concept of `port health'. Researchers have argued, for instance, that concern about tuberculosis in migrants in the 1950s and 1960s was tempered with relief that they did not integrate and spread the disease to the local population. It became a `disease of immigrants' and there was little acknowledgement that the incidence could be reduced by public health measures or by improvements in nutrition, housing, and the standard of living. Donaldson and Parsons for example, argue that Asian health care was dominated, in the 1960s, by ideas about `port medicine', where the emphasis was on exotic diseases, the danger of importing illness and the need to prevent its spread to the host population (1990: 83±4). Similarly, Ahmad claims that the emphasis on X-ray examinations at ports of entry, along with concerns about migrants generally, became `essential weapons for supporting and legitimating racist immigration policies' (1993: 20). Smaje has concluded that while there is a very high relative risk of tuberculosis

3 860 John Welshman among Asians, it is often forgotten that it is not necessarily a major health problem overall, accounting for only two per cent of deaths among women born in the Indian subcontinent. Rather, the concern with tuberculosis has often been negatively associated with fears of infection from abroad and with a restrictive view of ethnic health issues, associated with `port health' thinking (1995: 77±8, 117). As Smaje and Nazroo have argued, earlier work has tended to juxtapose cultural against socio-economic explanations of health and illness. The emphasis on port health has gone hand-in-hand with an assumption that material explanations of the ethnic patterning of health have been ignored. It is suggested that recent work has demonstrated that although rates of pulmonary tuberculosis are considerably higher among migrant groups than the indigenous population, this is not the result of individuals bringing in the disease themselves, but because they have acquired it in poor and overcrowded housing conditions in the inner cities (Donovan 1984: 665). Similarly, Ahmad and colleagues have written of tuberculosis that the `role of socio-economic and environmental factors... has largely been ignored' (1989: 49±50). These early studies quoted approvingly from epidemiological research that claimed to show that most migrants developed tuberculosis after entry, or demonstrated the importance of socio-economic factors in particular cities (Khogali 1979, Froggatt 1985). Overall, the tendency of social scientists has been to argue that although there was evidence that migrants had not brought the disease with them, but had acquired it through living in poor housing conditions, socio-economic factors were ignored in most studies. Rather, the emphasis of policy was always on port health and on providing medical examinations for migrants at ports of entry. Earlier research has been helpful in demonstrating the crude nature of such terms as `Asian' and the need for more accuracy in defining ethnic minority communities. In a review of research, Ahmad et al. have argued that the term `Asian' is inappropriate since it disguises the variations of language, cultural traditions, religion, national origins and class that exist between different groups. Moreover, they suggest that the research literature is heavily dependent on easily-available morbidity and mortality data, including those on rickets, tuberculosis, mental disorders and the health of mothers and children. They claim there has been little attempt to examine the use of general practitioner services, or to relate evidence on health and illness to the wider socio-economic background of high unemployment, poor housing and low educational status. The role of racism in health service provision and in the employment of black people has been ignored. In particular, they argue that much of the literature implies that the variations in health status between the Asian and White populations are the result of linguistic and, particularly, cultural differences (Ahmad et al. 1989). However, in the specific case of tuberculosis, this field of research is complicated by the complex nature of the disease. Recent debates in the biomedical literature about the relative influences of ethnicity and of socio-

4 Tuberculosis and ethnicity in England and Wales, 1950 ± economic deprivation have been inconclusive. Many studies, including some based on Liverpool, have confirmed that tuberculosis remains a disease of socio-economic deprivation (Spence et al. 1993). Yet a study of 32 London boroughs correlated average levels of notifications with overcrowding and the proportion of migrants, but not with unemployment or social class (Mangtani et al. 1995). Given these problems, some have concluded that disentangling the effects of deprivation from those belonging to an ethnic minority is almost impossible (Darbyshire 1995). Indeed, one recent study has suggested that, although a strong association exists between poverty and tuberculosis among white people, no such association exists among Asians, indicating ethnic differences in its epidemiology (Hawker et al. 1999). The reliance in that paper on area level rather than individual indicators of deprivation, illustrates some of the problems of finding appropriate indicators of socio-economic position for ethnic minority groups. More generally, it is clear that additional research in this field is required. Furthermore, the relative influences of importation and social deprivation are very difficult to study. The natural history of the disease (potentially very long incubation periods, subclinical infection that is difficult to identify except through tuberculin testing and difficulty in distinguishing primary disease from post-primary or reactivated disease) complicates such research (Davies et al. 1996). The close correlation between social deprivation and the area of residence of migrants makes it difficult to untangle the relative contribution of each factor. An analysis of tuberculosis transmission between nationalities in the Netherlands suggested that DNA fingerprinting techniques may offer one way forward (Borgdorff et al. 1998). However, even this method is not without its problems. A cautious interpretation would be to suggest that it is artificial to draw a distinction between importation and socio-economic effects. It is more likely that they both operate and do so synergistically. In fact, despite some interest in migration, the history of the changing health status of ethnic minority groups and the provision of services for them remains to be written (Marks and Worboys 1997). It is only recently for example, that historians have begun to examine the links between race, science and medicine (Ernst and Harris 1999). Given the complex nature of the relative influences of importation and social deprivation, this article does not attempt to untangle them. Rather it considers the way that the inferences drawn from the available research literature on tuberculosis among Irish and `Asian' migrants shaped the arguments of researchers and policy makers in the immediate postwar period. It looks at three separate but related areas ± the development of the research literature in the medical journals; the policy response by politicians and civil servants; and at service delivery through local public health departments. The focus of the article is England and Wales, but in the case of service provision, I also look briefly at the Midlands city of Leicester. I argue that the story was more complex than previous researchers have implied. The fear that tuberculosis was being

5 862 John Welshman imported by migrants was certainly a central concern of both early researchers and the medical establishment. Some papers, however, also reflect an interest in material explanations and in the roles of housing and work patterns in the transmission of the disease. A system of medical examinations at the ports of entry was not in fact implemented and it was at the local level that a system of surveillance was set up. Finally, despite much debate about the susceptibility of migrants, `racial' concerns were less evident than might have been expected. Tuberculosis and migration: the research literature Despite a history of arrivals and settlers stretching back hundreds of years, Britain has been comparatively late in coming to terms with the various health challenges that this has presented. The United States provides an interesting comparison. On the one hand, there is evidence that patients with tuberculosis, who were often migrants, tended to be heavily stigmatised. In the 1930s, for example, public health authorities wanted to ban interstate travel by tuberculosis patients and they were often turned away by the owners of boarding houses and hotels (Rothman 1994: 190±1). However, in the event, there was no ban on interstate travel and large-scale immigration at the end of the 19th century had been followed in the early 1900s by research on tuberculosis among groups like the Italians (Kraut 1997). In some respects, the research concentrated on individual habits, but it also focused on housing and employment conditions. The American experience of immigration, its timing and scale, provides an interesting contrast to that of Britain. Here there are few mentions of tuberculosis and migration in the medical journals before the 1950s. Even then, interest focused chiefly on London and on the Irish as the longest established migrant group (Greenslade et al. 1997). Although Britain was comparatively late in considering migrant experiences of health and illness, a body of research literature on tuberculosis grew gradually from the late 1950s. It was generated by the staff of chest clinics and public health departments in those cities, such as Birmingham and Bradford, that attracted sizeable migrant groups. The high incidence of pulmonary tuberculosis among those migrants deemed `Asian' was well known. In 1973, for instance, the British Thoracic and Tuberculosis Association showed that between 1965 and 1971, there had been a decrease of 43 per cent in notifications among people born in the British Isles, but an increase of 68 per cent among those born in India, Pakistan and the New Commonwealth. At the latter date, the five per cent of the population not born in the British Isles contributed 32 per cent of the notifications (British Thoracic and Tuberculosis Association Research Committee 1973). This earlier literature has largely been forgotten. Here I review it in order to examine the extent to which it concentrated on port health, how far it

6 Tuberculosis and ethnicity in England and Wales, 1950 ± embodied ideas about susceptibility and to see whether it made reference to tuberculosis as a disease of socio-economic deprivation. Tuberculosis as a port health issue It is certainly the case that many of these researchers argued that tuberculosis was imported by migrants, and that its rising incidence could best be tackled by medical examinations at the ports of entry. The early surveys concentrated on the experience of Irish migrants, but later research in the late 1950s and early 1960s focused on the experience of their Asian counterparts. One influential survey, by staff in the Department of Social Medicine at the University of Birmingham and based on the experience of the local Chest Clinic, confirmed that the incidence of tuberculosis among Irish and Asian migrants was high. From notifications in 1956 and 1957, it was concluded that for the Irish these were twice as numerous as might be expected, and for the Asian born, four to six times. In the case of the Irish, it was concluded that this was a susceptible non-infected rural population moving into an infectious urban environment. Given the high incidence of tuberculosis in parts of India and Pakistan, on the other hand, the excess notifications in the Asian group were `due more to the migration to this country of individuals already with tuberculous than to infection of susceptibles after arrival'. Thus, it followed from this analysis, that the recommended measures were BCG vaccination before departure for Irish immigrants and X-ray examinations before entry to Britain for migrants from India and Pakistan (Springett et al. 1958). Other studies were concerned with particular occupational groups, such as workers in the catering trade. A study of tuberculosis in the catering trade in Soho, published in 1961, was carried out by researchers from a London hospital chest clinic and mass radiography service. It was based on 47 pubs, along with 193 restaurants, cafes and coffee bars. Peter Emerson and colleagues found from X-rays of 2,611 employees that there were 8.0 `active' cases of tuberculosis per 1,000 workers, compared to 1.8 in the general population. Tuberculosis was four times more common, especially among those serving alcohol, preparing food and working in kitchens. Those deemed Chinese (mainly from Hong Kong) had the highest prevalence, at 53.8 cases per 1,000. Emerson et al. estimated from 86 cases with past or present tuberculosis that half had the disease before joining the catering trade (those from Hong Kong, Italy and Ireland), while others joined healthy and then developed it (those from Britain and Cyprus). They recommended that all new entrants to the trade should be X-rayed and tuberculin-tested, with BCG vaccination where necessary. To an extent therefore, Emerson et al. endorsed the value of routine radiographic examinations, noting that this would be particularly useful for the Chinese (Emerson et al. 1961).

7 864 John Welshman While the Birmingham research had concentrated on the experience of Indian migrants and that in Soho on the Chinese, work based in Bradford considered the case of arrivals from Pakistan. Here, D. K. Stevenson, a consultant chest physician, noted that although Asians with tuberculosis were first seen at the local Chest Clinic in 1954, by 1961, 127 of the 313 new notified cases of tuberculosis were from this ethnic group. Estimating that the Pakistani population of Bradford was around 7,000 in June 1961, Stevenson calculated the annual rate of incidence as 1.8 per cent, or 30 times greater than that of the `British' population. With regard to the new patients at the Chest Clinic, around one in five were Pakistanis (956 out of a total of 5,160). Stevenson also attempted to assess the usefulness of X-ray examinations at the ports of entry through an analysis of 131 Pakistanis with pulmonary tuberculosis seen at the local Middleton Hospital in the period 1955±59. Of these, 95 were thought to have had a disease of fairly recent onset, and 36 had been diagnosed with an obvious relapse of a previous disease. Stevenson estimated that 64 of these cases would have been recalled for investigation after X-ray examination ± around 40±50 per cent of the cases would have been picked up. Overall he endorsed the recommendation of the British Medical Association (BMA) that migrants should have X-ray examinations of the chest at the ports of entry (Stevenson 1962). Later work by Dr William Edgar, the former deputy Medical Officer of Health (MOH) and deputy School Medical Officer (SMO) for Bradford, built on this early research by Stevenson and had a similar emphasis. In 1963, for instance, when Pakistanis comprised 10,863 of the estimated 12,688 migrants in the city (86 per cent), `Asians' made up 212 of the 370 cases notified as suffering from tuberculosis (57 per cent). Control measures had included attempts to reduce overcrowding, selective use of mass miniature radiography and comprehensive tuberculin testing. Edgar argued, however, that since most of the migrants were single men, worked in hot humid conditions, spent their free time together and had poor diets (even though good meals were allegedly provided in the factories), the emphasis on the relief of overcrowding was of limited value. He concluded that if migrants continued to arrive without medical checks, the most elaborate systems after entry could not provide a solution, and some central form of control was necessary (Edgar 1964). With his earlier work on Birmingham, V. H. Springett (1964) had emerged as one of the leading authorities on tuberculosis in migrants, and he maintained his emphasis on port health measures in a series of publications. In some respects, his emphasis on the high incidence of tuberculosis in the countries of origin was supported by contemporary research carried out in India (Shah et al. 1960). But Springett's work also provided crucial support for those who were chiefly concerned with importation. In 1964, for example, Springett conceded that the high incidence among migrants was exacerbated by poor living conditions, but argued it was only through procedures such as chest X-rays at entry that there was any prospect of

8 Tuberculosis and ethnicity in England and Wales, 1950 ± controlling and eradicating the disease. He wrote that `it is in some ways fortunate that the immigrant groups with the higher tuberculosis rates ± that is, those from Asia ± have in general shown little tendency to integrate fully with other groups resident in the city'. The main aim was to `protect' those already in Britain (Springett 1964: 1094). Springett maintained that the high incidence among Asian migrants was due `mainly to their bringing with them the high rates they would experience in their own country', and suggested that X-rays on arrival with subsequent treatment would halve the problem (Springett 1966: 60). As late as 1971, in a review of tuberculosis control in Britain between 1945 and 1970, Springett continued to emphasise the importance of facilities for diagnosis and examination of contacts and appeared to place little stress on prevention through improving housing and working conditions (Springett 1971). Overall, this research had both highlighted the high incidence of tuberculosis among Asian migrants and endorsed the belief that it could best be controlled through a system of medical examinations at the ports of entry. In 1964, for example, the journal Tubercle concluded that the high incidence of tuberculosis among migrants was largely a reflection of the situation in their home countries ± the Chinese and Pakistanis probably had the disease at entry, while the West Indians developed it after arrival. The journal advocated a three-pronged attack ± efficient treatment of all cases, detection of established cases at the time of entry and measures to control tuberculosis in the countries of origin (Tubercle 1964). If much emphasis was placed on medical examinations at the ports of entry, there was also evidence of an unsympathetic approach to the health of migrant groups in general. Doctors from Birmingham and Bradford, for instance, continued to focus on such issues as venereal disease, mental illness, worms, illegitimacy and high infant mortality, and seemed to have little sense of socio-economic deprivation as a factor in health status. In the case of tuberculosis, it was concluded that `in the absence of screening procedures, it is inevitable that much tuberculosis will be brought in by the migrants' (Proceedings of the Royal Society of Medicine 1968: 23). Tuberculosis as a disease of socio-economic deprivation But although some researchers did place much emphasis on the value of medical examinations at ports of entry, others continued to look at tuberculosis as a disease of socio-economic deprivation. This was to be expected, given what was known about tuberculosis, but it also incorporated ideas about the special susceptibility of migrants to the disease. As already noted, the early studies were concerned with Irish migrants, while later research concentrated on their Asian counterparts. But although there were some similarities in the way that the problem was constructed in the case of these ethnic groups, there were also important differences. Moreover, even in this

9 866 John Welshman early period, some already argued against the emphasis on racial characteristics. A writer in the Lancet, for instance, dismissed this explanation, noting of Irish migrants that `the men commonly engage in heavy manual work, the women often become waitresses or maids and because they have no home life, they eat irregular meals and often poorly nutritious food' (Lancet 1954: 1282). This writer clearly acknowledged the importance of socio-economic deprivation, his reliance on cultural stereotypes notwithstanding. Empirical investigation suggests it is this dimension of the story and these ambiguities and complexities, that have been lost through the reliance on the concept of port health. One of the first reports was produced by Evelyn Hess and Norman Macdonald in 1954, based on their experiences in the North West Metropolitan Regional Hospital Board. They noted that earlier work showed a higher mortality from tuberculosis among Irish migrants and set out to investigate the epidemiological implications of this `racial or ethnic susceptibility'. They reported the results of a study based on 292 patients with pulmonary tuberculosis in five hospitals in North London and Hertfordshire. Of these, 36 per cent had been born in Ireland, 44 per cent in Greater London of London-born parents and 20 per cent had Irish parents or grandparents. They estimated that the proportion of Irish patients was high ± at least three times that expected on the basis of the relative numbers at risk. Giving examples of migrants arriving in cities such as London, Calcutta, and Cairo, Hess and Macdonald suggested the same basic pattern ± `low degree of inherited resistance, high degree of tuberculinnegativity, susceptible age-group, poor standard of living ± providing a distressingly suitable human culture medium for the waiting bacillus' (1954: 136). In fact, while they highlighted the significance of a family history of tuberculosis, they rejected a racial explanation in favour of one based on the `susceptibility' of rural migrants. Hess and Macdonald concluded that `migration of the immunologically ill-equipped descendant of rural stock', often young and tuberculin-negative, to cities where the risk of infection was greater, was one of the ways in which tuberculosis spread (1954: 137). They recommended BCG vaccination of immigrants before they left Ireland and follow-up in Britain, but noted that this was not a substitute for good nutrition, improved housing or adequate treatment of infectious patients (Hess and Macdonald 1954: 137). Other reports began to compare the experience of the Irish with that of other migrant groups. The report by Hess and Macdonald was followed by another published in 1958 by G. Z. Brett, who worked in the same area but in the mass miniature radiography service. He examined the records for 1956 of a mass radiography unit whose catchment area covered the London boroughs of Islington and St Pancras. Of a total of 32,228 examinees, 27,655 were born in the United Kingdom, 1,801 were Irish, 475 Cypriot and 978 West Indian. Brett reported that the prevalence of tuberculosis among Irish migrants was between three and seven times that of the control group and

10 Tuberculosis and ethnicity in England and Wales, 1950 ± was also higher in Cypriots, though not among West Indians (1958: 28). Brett suggested that his findings not only supported the contention of Hess and Macdonald, but indicated that their ratio of 3:1 was a conservative estimate. On the question of whether Irish migrants brought the disease with them or contracted it in Britain, Brett was cautious, noting that at all stages of stay the Irish in the sample had a higher rate of active, as well as infectious, disease than the indigenous population (1958: 26). Given Britain's earlier history of immigration, the term `indigenous' proved surprisingly long-lived in these debates. While the work conducted by V. H. Springett and others based on Birmingham had largely endorsed the concern with `port health' measures, this emphasis was contested by other researchers in the West Midlands. Some of this work was not conducted by the personnel of chest clinics, but by the staff of local public health departments and their methods and findings reflected their different professional perspective. One survey, conducted in July 1956 by J. F. Skone and S. Cayton, the MOH and Chief Public Health Inspector for West Bromwich, confirmed that new arrivals tended to live in overcrowded conditions. They found that 838 West Indian, Indian and Pakistani migrants in West Bromwich were living in 72 houses, and over half were living two or more to a room. Communal living was a financial necessity ± language, `custom' and circumstances had brought them together initially, and poverty meant they could not move (1957: 122±3). Skone and Cayton suggested that their findings did not support the allegation that migrants arrived in Britain suffering from tuberculosis ± instead the results indicated `that they contract the disease readily when they work in heavy industry and live in overcrowded conditions' (1957: 125). Other surveys linked the high incidence of pulmonary tuberculosis among Indian migrants to their having developed the disease after arrival, and stressed the need for co-operation between general practitioners, public health departments, factory medical officers and chest clinics. Roe (1959), based at the Uxbridge Chest Clinic, estimated that of a group of 35 Indian patients with tuberculosis 12 probably had the disease on entry, two had possibly acquired the disease in Britain, and 21 had definitely developed the disease after arrival (1959: 387). In Birmingham, Springett had found that the disease was predominantly chronic, claiming that this indicated the patients already had tuberculosis on arrival in the country. However, Roe argued that the Uxbridge evidence suggested many of the Indian immigrants had acquired tuberculosis in Britain, since many had primary disease, pleural effusion and miliary tuberculosis of the lungs. In fact, this survey was based on very small numbers and was highly flawed since, as already suggested, there is no clinically reliable way of distinguishing between primary, post-primary and reactivated disease. Nevertheless, the inference drawn by the author from this evidence was that health professionals, whether in public health or general practice, should combine to provide X-rays after arrival, tuberculin testing and BCG vaccination, offer health

11 868 John Welshman education in conjunction with local Indian Workers' Associations and attempt to reduce overcrowding (Roe 1959). The Roe Study was based on the London suburb of Southall and, as immigration increased, the experience of other provincial towns and cities began to attract attention. In Wolverhampton for example, Aspin (1962) found from new cases notified in 1960 that there were four times as many among Indian migrants as might be expected in a similar number of local inhabitants. He claimed that of 67 new cases notified in Wolverhampton since 1954, only 13 (19 per cent) would have been picked up by chest radiography at the time of entry. It was suggested the rest had developed the disease in England, even though their housing and living conditions, though poorer than those of the host community, were not much worse than in their countries of origin (1962: 1387). Aspin recommended that all Indians should have X-rays after entry and before being accepted on general practitioner lists, and all adults should have annual check-ups (Aspin 1962). These findings were supported by those of John Corbett, a general practitioner in Wellingborough, who observed the high incidence of tuberculosis among a small group of Indian immigrants who had migrated there from the State of Bombay. He noted that all the rooms in these houses were converted into bedrooms for single men or for married couples with children. Overall, he concluded that while climate, working conditions and poor diet, were all factors in the high incidence of tuberculosis, `overcrowded living conditions, with consequent possibilities for infection' was probably the most important cause (Corbett 1961: 332). Moreover, even those researchers whose studies had endorsed the emphasis on medical examinations at ports of entry also showed a sympathetic appreciation of the links between tuberculosis and socio-economic deprivation. This was certainly the case with the work in Bradford. Even though Stevenson had stressed the value of `port health' measures, he observed that the migrants from Pakistan worked in such occupations as the textile trade, engineering industry, and public transport, and noted that they lived in the older, central wards of the city, which tended to be overcrowded. He claimed from the in-patient survey of hospital cases that while 40 to 50 per cent of the cases would have had an abnormal X-ray picture at the time of immigration, at least 50 per cent of those eventually developing a pulmonary infection probably acquired it in Britain. Thus, he argued, there were two sides to the issue ± it was both a question of `the tuberculous immigrant' and the `susceptible Pakistani' (Stevenson 1962: 1385). His colleague Dr William Edgar agreed that the arrivals in Bradford from Pakistan worked in the local textile trade, foundry work, public transport, and unskilled engineering, and he not only advocated the selective use of mass miniature radiography, and tuberculin-testing, but also supported attempts to reduce overcrowding (Edgar 1964). With hindsight, it can be seen that many of these studies did not have enough data to make sensible conclusions about the relative influences of

12 Tuberculosis and ethnicity in England and Wales, 1950 ± social deprivation and importation. Nonetheless, what is perhaps more important in terms of the development of health policy is not the scientific accuracy of the conclusions drawn, but the judgements made on the basis of this admittedly flawed evidence. Clearly, therefore, it is the case that many researchers here argued that the perceived high prevalence of tuberculosis could best be reduced by X-ray examinations at the ports of entry. When socio-economic issues were discussed, this was very much subordinate to a discussion of importation. But, an interesting distinction was drawn between the Irish and Asian migrants in this respect ± whereas the former were susceptible in the new urban environment, the latter were perceived as bringing tuberculosis with them. Thus, attitudes towards tuberculosis were both shaped by notions of race and influenced by ideas about the rural/ urban divide. Moreover, some researchers were willing to examine the socioeconomic background to tuberculosis, particularly overcrowding and housing conditions and, less so, nutrition. Although it might be expected that the staff of chest clinics would place more emphasis on `port health' and those in public health departments focus on housing and employment, this neat dichotomy did not work out in practice. These different interpretations of the evidence were reflected in the policy response, both by central government departments and at the local authority level. The policy response: central government departments It is important to consider how policy makers come to decisions based on evidence that with hindsight can be seen as inadequate, or on an uncritical and biased review of the literature. The research literature on tuberculosis and immigration did not develop in a vacuum, but was itself a product of a specific historical period when various policy developments revealed wider social and political influences. In particular, much of the research published in the 1960s, reflected pressure on the part of the BMA that all migrants should have medical examinations at the various ports of entry to the United Kingdom. Here, I discuss in more detail the policy response to tuberculosis and ethnicity, concentrating on the BMA and the relevant central government departments, the Ministry of Health and its successor the Department of Health and Social Security (DHSS). Particular use is made of the annual reports of the Chief Medical Officer and of internal departmental files, along with contemporary medical journals and newspapers. The focus is not so much on policy in a party-political sense, but rather on the way in which it was interpreted at a departmental level. At first glance, it would appear that the policy response was dominated by `port health' measures and by an insistence on the need for medical examinations at the ports of entry. Contemporary newspapers indicate that there certainly was evidence of a `moral panic' on the question of migration and tuberculosis on the part of some journalists and social groups (Daily

13 870 John Welshman Herald 1953). Moreover, discussions within central government departments were revealing of the ways in which participants viewed issues of race and immigration. On the Irish, one civil servant at the Ministry of Health wrote in October 1955 that `the type of immigrant chiefly involved is of comparatively low intelligence & particularly wayward in habit' ± he doubted whether any health advice would be heeded. There was a conjunction in the representation of the tuberculosis patient with a kind of early `underclass' concept. The same civil servant asked whether the Irish migrant `represents that stratum of society in this country to which the impetus of our anti-tuberculosis drive ± by local authorities & by radiological units ± must increasingly be directed?' 1 In these ways, there were important continuities in the ways that ideas of race shaped attitudes towards both Irish and Asian migrants. Certainly, the perceived need for medical examinations was the subject of consecutive recommendations by the Tuberculosis and Diseases of the Chest Group of the BMA, and by other pressure groups. In November 1956, for instance, the Health Committee of the Association of Municipal Corporations recommended that all migrants to Britain should be medically examined in their own countries. It was suggested that those suffering from tuberculosis, mental illness and deficiency, and other infectious diseases should be excluded. Similarly, the Standing Tuberculosis Advisory Committee of the Central Health Services Council stated in 1959 that only a system of strict medical control of immigration would be completely effective in preventing the spread of tuberculosis (Skone 1962). The extent to which the BMA colluded with this moral panic in the early 1960s can be seen in leader articles and correspondence in the British Medical Journal. In September 1961, for example, the BMA's Representative Body stated that it viewed with concern the admission of migrants without medical checks and argued that all should have chest X-rays. The BMA's Council subsequently passed three resolutions to this effect in December (BMJ Editorial 1961). Similarly, in 1962 a leader article concluded from the recent research evidence, that `Asian immigrants, and especially Pakistanis, constitute a source of tuberculous disease and infection out of proportion to their numbers' (BMJ Editorial 1962). At the same time, there were signs that a pathological view was taken not just of migrants with tuberculosis, but of all patients with the disease. One correspondent in the journal wrote that the need for compulsory treatment was exemplified by `the hard core of our own recalcitrant and often vagrant tuberculous' (BMJ 1962: 50±1). A further development was the setting up by the BMA, in January 1965, of a small working party to investigate the health `problems' of migrants. Its composition was revealing of the wider intellectual climate ± chaired by Dr C. Metcalfe Brown, the MOH for Manchester, it included a senior airport/ seaport medical officer, chest physician, and radiologist. This produced an interim report in July and a final report in December. Interestingly, the

14 Tuberculosis and ethnicity in England and Wales, 1950 ± working party drew up a long list of those diseases that should be excluded and these included yaws, leprosy, yellow fever and dysentery, along with mental disorder, drug addiction and alcoholism. The list revealed the stress placed on the threat to others, rather than the health of the migrant and also embodied a pathological view of the patient. The main recommendation of the working party was that all migrants should have medical examinations in their country of origin, with further examinations and follow-up measures after entry, `in view of the low natural immunity to disease of many immigrants and the social conditions under which they live' (BMJ 1965: 1424). Some of the civil servants at the Ministry of Health supported the BMA and the idea of X-raying migrants on arrival in Britain. George Godber, for example, reported in May 1962 that in Geneva he had observed 650 workers being X-rayed in five-and-a-half hours. He concluded that Switzerland had shown what could be done and, that while the situation in Britain was more complex in that immigration was on a different scale, it would not be impossible to do likewise if the financial costs could be met. 2 The 1962 Commonwealth Immigrants Act stated that refusal of entry could be on grounds of health, criminal record, security, or previous deportation. In the case of health, this was as if it appeared to immigration or medical officers that the settler was `a person suffering from mental disorder, or that it is otherwise undesirable for medical reasons that he should be admitted'. Some measures were taken, including the setting up in February 1965 of an experimental X-ray machine at Heathrow Airport. Furthermore, in 1969 arrangements for medical examination before departure were extended to cover `non-entitled' dependants; medical attention at airports and ports concentrated on `entitled' dependants (wives and young children) who often had not had a medical examination. Despite this evidence, however, central government was forced to opt for a simpler policy that was still essentially concerned with surveillance, but which moved its site from the ports of entry to those local authorities that received large numbers of migrants. It was not based primarily on medical examinations, but on a system whereby port medical officers forwarded the addresses of recently-arrived people to local MOsH in the cities where they were planning to settle. They were to advise them to register with a family doctor, and to provide other services for both children and adults, including tuberculin testing and BCG vaccination (Ministry of Health 1966: 27). It was clear that this system did not operate effectively. Figures for 1967, for example, revealed that while addresses for 35,985 migrants were sent to local authorities, only 23,541 (65 per cent) were visited by the staff of local public health departments (DHSS 1968: 79±80). Yet, despite this evidence, the principle that migrants were allowed to enter the country as long as they reported to a MOH was reaffirmed in the 1968 Commonwealth Immigrants Act. A range of factors were involved in the adoption of this policy. One was the advantages of relatively open borders to a government concerned about

15 872 John Welshman a growing economy in which the demand for labour outstripped supply. A second was wider political sensitivity by the Labour government to the whole issue of immigration. A third was the practical difficulties involved in attempting to X-ray large numbers of people and the linguistic and other administrative problems that this would have created. Certainly, it is clear that the pressure from the BMA and other groups in favour of medical examinations met continued resistance from the Ministry of Health over our entire period, from the mid-1950s to the early 1970s. As early as July 1955, for example, Iain Macleod, Minister for Health, had stated that tuberculosis among Irish migrants was not sufficiently serious to justify health checks at the ports of entry. 3 Moreover the Ministry's opposition to compulsory medical examinations led to serious friction with the BMA. In December 1961, for instance, the BMJ stated that a letter from the Ministry, written in response to the recent BMA Council resolutions, `beats all records for ministerial evasion' (BMJ Editorial 1961). The journal clearly felt that the system announced by the Ministry in January 1965, that relied on notifying addresses to local MOsH, did not go far enough. Indeed the BMJ compared the failure to take action on medical examinations in the early 1960s, to the delay in taking up diphtheria immunisation in the 1930s. In the opinion of the BMA, on both diphtheria immunisation and medical examination, Canada had been the exemplar (BMJ Editorial 1965). Exactly why the BMA took this line, and whether it was a question of personalities or professional interests, is not clear. Probably, it perceived that in the absence of an efficient screening system it was general practitioners who would be left to pick up the pieces. But the emphasis on medical examinations at the ports of entry must also be seen in terms of its wider symbolic value, the limited (and acknowledged) effectiveness of the measure notwithstanding. Governments, both Labour and Conservative, opposed the BMA proposals for medical examinations on entry and set up a simpler system of surveillance at the local level. In fact, Ministry of Health officials showed a persistent resistance to the idea that tuberculosis was coming into the country with new arrivals, relying instead on the well-worn theme of the `susceptible' migrant. In January 1956 for example, one civil servant opposed medical examinations, writing that the consensus of opinion was that the real problem was `the susceptible people who come for the first time in contact with the stresses and risks of town life here rather than those entering the country in an infectious condition'. 4 Edith Pitt, the Ministry's Parliamentary Secretary, stated in the Commons in December 1961 that `she had no reason to believe that immigrants from the Commonwealth had been responsible for bringing infectious disease to this country to an extent likely to involve risk to public health' (BMJ 1961: 1720). Similarly the Permanent Secretary of the Ministry of Health conceded in March 1962 that Asian immigrants living in poor housing conditions had a higher incidence of tuberculosis than the general population. However, he noted `whether the

16 Tuberculosis and ethnicity in England and Wales, 1950 ± incidence is much higher than it would be among UK natives living in similar social and economic conditions is not certain'. 5 It might be suggested that civil servants stuck to this line simply because they opposed the emphasis on medical examinations at the ports of entry. But reports by other advisory bodies and individuals also showed an awareness of the links between the high incidence of tuberculosis among migrants and socio-economic deprivation. The Standing Medical Advisory Council for instance, noted in a report on the changing epidemiological profile of tuberculosis that some cases among migrants could be missed by X-ray examinations at the ports, and other `susceptible groups' might acquire the disease after entry (Standing Medical Advisory Council 1964: 1). His earlier comments on X-ray examinations in Geneva notwithstanding, a similar emphasis was apparent in the annual reports of George Godber, the Chief Medical Officer. In his report for 1966, for example, he wrote that `as many Commonwealth immigrants in their first years in this country tend to live in overcrowded conditions and to be among the lower income groups, the risk of spread of the disease is considerable' (Ministry of Health 1967: 68±70). Furthermore, in 1969 the CMO observed of migrants that `it is not their importation of infection but development of clinical tuberculosis after arrival that is the chief cause of concern' (DHSS 1970: 54). However, despite this recognition that socio-economic issues were important in the transmission of the disease, little was done, and the policy response centred on increased surveillance at the local level. The local response: tuberculosis and migration in Leicester In addition to the research literature on immigration and tuberculosis and the policy response by central government departments, a third area for investigation is that of the provision made by local public health departments. Here, I focus on their response through a case-study of one local authority, the Midlands city of Leicester. Leicester had a growing migrant population from the 1960s and provides an opportunity to test some of the arguments that have been made about local health departments and their response to immigration. Catherine Jones (1977), for instance, has provided an essentially optimistic interpretation, suggesting that health departments in general were more efficient than other branches of the social services in their provision of health education and attempts to tackle problems of literacy. This included special training for health visitors and district nurses and moves towards employing staff from New Commonwealth countries (1977: 198±205). Yet it has also been argued of Leicester that immigration posed a serious test to the public health services in the city, and that the statistics in the local MOH reports on the incidence of tuberculosis among minority groups were inaccurate and alarmist (Simmons 1974: 105±6). A local study enables us to explore the key question of