Tuberculosis: old reasons for a new increase?BMJ 1995; 310 doi: https://doi.org/10.1136/bmj.310.6985.954 (Published 15 April 1995) Cite this as: BMJ 1995;310:954
- Janet H Darbyshire
Socioeconomic deprivation threatens tuberculosis control
Notifications of tuberculosis have increased in England and Wales over the past few years, as in other European countries and the United States.1 2 3 An estimated 8000 extra cases occurred between 1982 and 1993 in England and Wales, but the 95% confidence interval is wide (3000 to 12000).1 At least part of the increase may be an artefact—for example, the creation of consultants in communicable disease control in 1988, together with local initiatives (such as that described by Brown and colleagues (p 974),4 may have resulted in a substantial fall in the undernotification previously reported in several areas.5 The increase in notifications has been largely for non-respiratory tuberculosis,1 in which the new consultants may have had their biggest impact—undernotification is more likely in specialties other than respiratory medicine. On the other hand, evidence exists that undernotification of tuberculosis, particularly in association with HIV infection, is still common.6
Factors contributing to a real increase are likely to be multiple and may vary among areas and populations. Notifications of tuberculosis in Britain fell steadily long before specific chemotherapy was available. It was recognised in 1899 that “the most powerful factors in producing tuberculosis are—(1) air contaminated by the so-called tubercle bacillus, (2) food inadequate in purity, quality and quantity, (3) confined and overcrowded dwellings, (4) a low state of general health and resisting power of the body.”7 The fall was attributed primarily to improved socioeconomic conditions and the isolation of infectious cases. Temporary increases in tuberculosis associated with wars were explained by poor nutrition, overcrowding, and fewer beds in sanatoriums.8 The continued fall after effective treatment was introduced was slowed but not reversed by the arrival of immigrants from countries with a high prevalence of tuberculosis.9 Much higher rates, particularly in the Indian, Pakistani, and Bangladeshi ethnic groups, have been documented on several occasions over the past 30 years.10 11 12 The increase in notifications since 1988 is of particular concern as it seems that immigration may not be the only factor and indeed may not be the most important one in some areas. The papers from Mangtani et al (p 963)13 and Bhatti et al (p 967)14 in this week's journal indicate that socioeconomic deprivation may also be important. Nevertheless, disentangling the effects of deprivation from those of belonging to an ethnic minority on the incidence of tuberculosis is almost impossible.
Unsurprisingly, in the 32 London boroughs tuberculosis is associated with unemployment and immigration; of more concern may be the association between recent increases in both tuberculosis and unemployment.13 In Britain the greatest increases in tuberculosis between 1980 and 1992 occurred in the poorest 10% of the population (on the basis of the Jarman index). In this group notifications increased by 35% compared with a national increase of 12%. Indeed, an increase occurred only in the poorest 30% of the population. The increase in the borough of Hackney (with a rate four times the national average and an increase from 172 cases in 1986-8 to 305 in 1991-3) was not limited to new immigrant groups and refugees, although they accounted for almost half of the excess cases.
Homelessness increases risk
Poverty, unemployment, and homelessness are inextricably linked, and all increase the risk of tuberculosis. Because of the difficulties of management, particularly in homeless people, drug resistance may become more common. Recent surveys of single homeless people in London carried out by Crisis showed that 2% of people living in hostels or using day centres had active tuberculosis.15 Of further concern is the increasing number of young homeless people, which is relevant to any future decisions about the BCG programme in England and Wales.
In the United States HIV infection is undoubtedly the most important cause of the increase in tuberculosis, but the breakdown of tuberculosis control programmes is a contributory factor. In Britain, HIV infection seems to have had a relatively small impact on tuberculosis.16 If the epidemic of HIV infection in India increases this will inevitably increase the overlap between HIV infection and tuberculosis in immigrant groups in Britain. Currently among immigrants such an overlap is limited largely to those from sub-Saharan Africa.
Since the introduction of the NHS internal market concern has been expressed that the systems for controlling tuberculosis, which have so far been largely successful, will not be adequately maintained at a time when they need to be strengthened. As Bhatti and colleagues comment, however, this may not be as important as reversing the underlying fall in real income among the poorest section of the population. The failure to reduce tuberculosis in most developing countries in spite of the availability of effective chemotherapy has been attributed to the failure to improve socioeconomic conditions, and the evidence from Britain supports this.