Search

Setting: The findings of the principal controlled trials of BCG vaccination against tuberculosis are reviewed., Objective: To investigate possible reasons for the wide variations in vaccine efficacy in different trials, and in particular for the apparent adverse effect in the first few years in the trial in Chingleput, South India., Observations: (1) A substantial reduction, or even a reversal, of overall efficacy could result if the assessment of vaccine efficacy included a subgroup adversely affected by vaccination. (2) Several trials include in the assessment many subjects with weak initial tuberculin sensitivity, due either to environmental mycobacterial infection or to Mycobacterium tuberculosis infection. It is accepted that vaccine efficacy may be moderately reduced in the former subgroup. It is postulated that the latter subgroup may be at risk of reactivation of tuberculosis soon after vaccination, perhaps from focal reactions due to enhancement of their weak sensitivity. (3) The low levels of efficacy in several trials, and the early adverse effect in Chingleput, are broadly consistent with this hypothesis., Conclusion: Clinical tuberculosis due to BCG vaccination of subjects with weak tuberculin sensitivity following M. tuberculosis infection may make an important contribution to the variations in efficacy found in clinical trials.

Between special surveys in 1971 and 1978/79 the estimated annual tuberculosis notification rate for males in England fell from 30.5 per 100 000 to 22.6 per 100 000 (a decline of 3.8% per year). For females the rate fell from 18.4 per 100 000 to 15.8 per 100 000 (1.9% per year). In each survey the lowest rates were those for the white ethnic group born in the United Kingdom. The highest rates, some more than 50 times as great, occurred in immigrants from the Indian sub-continent (Indian or Pakistani/Bangladeshi). The rates for immigrants from the West Indies were 3 to 4 times as great as than those for the white group. The most rapid reductions in rate between the surveys, of about 10% per year, occurred in West Indian immigrants of both sexes and in Pakistani/Bangladeshi male immigrants. The rate for Pakistani/Bangladeshi females fell by 6.5% per year. For whites born in the U.K. the annual rate of decline was 5.1% in each sex. There was very little change for Indian immigrants of either sex. Between the surveys, continued immigration of groups from the Indian subcontinent with high notification rates considerably slowed the decline in notification rate for the whole population. The steep downward trend in notification rate for the white ethnic group may be expected to continue, but changes in the other ethnic groups are more difficult to assess because they are influenced by so many uncertain factors. In addition, the trends in the non-white ethnic groups born in the United Kingdom cannot yet be ascertained, but will become of increasing importance.

The aims of this paper were to estimate the numbers of tuberculosis notifications in young white adults which will be prevented in the next 25 years by the schools BCG vaccination scheme, and to assess the numbers of additional notifications if the scheme were to be discontinued. Assuming that in the white ethnic group in England and Wales the decline in tuberculosis notification rates (8-10% per year for ages less than 45 years) and efficacy of BCG vaccination (75-80%) are maintained, it is estimated that the scheme for BCG vaccination of schoolchildren with its present coverage will prevent 217 notifications in those aged 15-29 years in 1993, 119 in 1998, and 69 in 2003. The epidemiological consequences of stopping the BCG in schools scheme, whenever this occurs, would be a substantial slowing of the rate of decline of tuberculosis notifications, confined almost entirely to the 15-29 years age group, for a period of about 15 years, after which the steeper decline would resume. If the scheme stopped at the end of 1991 the annual number of additional notifications would slowly increase to a maximum of just over 80 about 15 years later, and then decrease. For stopping at the end of 1996 the maximum annual number of additional notifications would be about 50.

The effectiveness of BCG vaccination, at about age 13 years, in the prevention of tuberculosis at ages 15-24 years in England and Wales in 1983, has been determined by the same method as in two previous surveys in 1973 and 1978. In 1983, the tuberculosis notification rate among those vaccinated in the schools' scheme was 3.30 per 100,000, compared with an estimated rate of 13.20 per 100,000 among those who were tuberculin negative and not vaccinated in the scheme. The protective effectiveness of BCG vaccination at ages 15-24 years in England and Wales in 1983 is thus estimated as 75%; the estimated efficacy in the white ethnic group is closely similar, namely 76%. The estimated efficacy of BCG at ages 15-19 and 20-24 years in the three surveys, both in the white ethnic group and in the entire cohort, has been compared with that found in the Medical Research Council's controlled trial of tuberculosis vaccines which began in 1950. There is no evidence of any decrease in the protective efficacy of BCG vaccination between the four cohorts of young adults, which span a total period of 29 years. However, there were steep decreases between the cohorts in the annual notification rates for the white ethnic group; these decreases occurred in the BCG vaccinated and in the tuberculin negative unvaccinated groups, as well as among those found tuberculin positive (and not vaccinated) in the schools' scheme. It is concluded that the level of protective efficacy of BCG vaccination at ages 15-24 years is high, and has remained unchanged since the start of the BCG in schools' scheme. However, as the tuberculosis notification rate in young adults has decreased steeply throughout this period, and is continuing to decrease, the benefits to be expected from the BCG in schools' scheme will decrease equally rapidly.

In two national surveys of tuberculosis notifications in England conducted in 1978/79 and 1983 the estimated annual notification rates for the Indian subcontinent (Indian, Pakistani, and Bangladeshi) ethnic groups were considerably higher than the rate for the white ethnic group. The mean annual decline in rates between the surveys appeared to be greater for the Indian and the Pakistani and Bangladeshi ethnic groups, 15% and 16% respectively, than for the white ethnic group (7%). Data from two small sample population surveys, the National Dwelling and Housing Survey in 1978 and the Labour Force Survey in 1983, were used to calculate the rates. However, comparison of the estimates for the population of Indian subcontinent ethnic origin in England from these surveys revealed discrepancies between them. Additional information from the Labour Force Survey on the year of first entry to the United Kingdom (UK) permitted the calculation of new estimates for the 1978 population, and based on these estimates the annual notification rates for 1978/79 were 287 per 100,000 for the Indian and 286 per 100,000 for the Pakistani and Bangladeshi ethnic groups. The rates for 1983 were 178 and 169 respectively, and the mean annual decline between the surveys was 11% for the Indian and 12% for the Pakistani and Bangladeshi ethnic groups. There were important changes in the characteristics of the population of Indian subcontinent ethnic origin in England between 1978 and 1983, and therefore the rates for both surveys have been standardised by the method of direct standardisation to a common reference population. Standardizing for year of entry to the UK, place of birth (UK or abroad), age, and sex reduced the mean annual decline in the notification rate to 4% for the Indian and 9% for the Pakistani and Bangladeshi ethnic groups. The much greater reduction in the rate of decline in the Indian ethnic group is due to the substantial decline between the surveys in the proportion of recent immigrants, the group with the highest annual notification rate, in that population. Future trends will continue to be influenced by immigration patterns, but it will also be important to monitor the rates among the increasing proportion of the population born in the UK or resident in England for more than five years.

Since the early 1960s notification rates for tuberculosis in England and Wales for the whole population have been influenced by high rates in certain ethnic groups. Using data based on country of birth from the British (Thoracic and) Tuberculosis Association surveys of 1965 and 1971, and based on ethnic origin from the Medical Research Council surveys in 1978/79 and 1983, rates for the white ethnic group have been estimated at those four times, and compared with the published rates for the whole population in 1953, when only a very small proportion was of non-white ethnic origin. Between 1953 and 1983 the notification rate for the white ethnic group fell from 122.2 to 11.3 per 100,000 for males, an annual decline of 7.7%, the corresponding rates for females being 90.1 and 5.8, an annual decline of 8.8%. The greatest annual declines occurred between 1953 and 1965, 9.4% for males and 11.2% for females. The annual declines in the most recent period, 1978/79 to 1983, were 6.9% for males and 7.3% for females. In both sexes the decline was greatest in the 15-24 year age group and least in the oldest age group, and this has led to a change in the age pattern of annual notification rates. The highest rates in both sexes occurred in young adults in 1953 but in the oldest age groups in 1983. There is however no evidence of any cohort experiencing an increase in notification rate with increasing age.

Analysis of the tuberculosis notifications and the statistics of B.C.G. vaccinations among 13-year-old schoolchildren in Birmingham, England, between 1953 and 1969 shows that the Copenhagen strain of B.C.G. (liquid batches issued 1953-61) and the Glaxo Laboratories strain (freeze-dried batches issued 1962-68) were of similar high efficacy in the prevention of tuberculosis. The method of analysis could be used for the surveillance of the protective efficacy of B.C.G. vaccine in other population groups where it is used routinely but not universally.