The history of tuberculosis in the United States

Although collection of cause-specific mortality data in the US began in 1900, data from all states were not included until 1933. A rate of 'new' cases of tuberculosis

FIG. 2. Funding by the Centers for Disease Control for tuberculosis control.

was determined annually beginning in 1930 but this figure did not become an accurate reflection of disease incidence until 1952 when inactive cases were no longer included. As shown in Fig. 1B, morbidity and mortality rates were consistently decreasing with the exceptions of increased death rates during World War I, a transient levelling of the new case rate in the early 1960s and an apparent increase in new case rates in the mid-1970s due to a change in reporting criteria. Although it is stated frequently that the tuberculosis situation was improving prior to the advent of chemotherapy, it is clear from Fig. 1B that the decline in mortality rates accelerated at the time chemotherapy became available in the late 1940s. Because accurate data on case rates weren't available until after chemotherapy appeared, its impact on morbidity can't be determined.

Although it has been the subject of considerable debate, the decline in tuberculosis mortality prior to the advent of chemotherapy has been generally attributed to improvements in socioeconomic conditions with less crowding and better nutrition. However, there are theoretical reasons, based on mathematical modelling, to think that the decline was in part a function of the natural dynamics of tuberculosis epidemics (Blower et al 1995). Regardless of the cause of the decrease in tuberculosis mortality and morbidity, the common assumptions were that the causative factors, be they biological or sociological, would be more or less constant, and would result in the ultimate disappearance of the disease (Frost 1937). The conditions that came to prevail in the 1970s, 1980s and 1990s in the US were not foreseen.

Not shown in Fig. 1B, but apparent in Fig. 1A, are the two subsequent deviations in the downward trend of morbidity: a levelling of the trend in 1979, 1980 and 1981 and, more notably, a levelling followed by an increase in 1984—1992. The first of these deviations was due to an influx of Southeast Asian refugees and was much more noticeable in data from specific areas, such as San Francisco. This increase in foreign-born cases was a portent of an important trend in tuberculosis epidemiology in the US and in western Europe: the globalization of tuberculosis (McKenna et al 1995, Raviglione et al 1995, Reider et al 1994).

A second factor that is evident is the increase in tuberculosis in persons with HIV infection. Fortunately, in San Francisco the incidence of AIDS has decreased substantially, as has the incidence of tuberculosis and the number of cases with HIV infection. However, HIV infection remains an extremely important risk factor for tuberculosis in most urban centres in the US and in the world (Hopewell 1997).

The second change in direction of the incidence curve of tuberculosis is the subject of this chapter and was a much more serious occurrence that highlighted fundamental flaws in the country's tuberculosis control programme. As can be seen in Fig. 2, targeted funding for tuberculosis at the federal level disappeared in 1972 and didn't reappear until 1978. Even after targeted funds were appropriated, funding levels remained at or below what they had been in the late 1940s. This is

FIG. 3. Initials

Per cent change in numbers of reported tuberculosis cases between 1984 and 1992. stand for individual states.

FIG. 3. Initials

Per cent change in numbers of reported tuberculosis cases between 1984 and 1992. stand for individual states.

not to say that there was no money for tuberculosis control: local funding continued, although probably at a reduced rate until it was evident that there was a serious problem in the late 1980s.

At least in part the reasons for the lack of broad concern with tuberculosis were related to the fact that the resurgence of the disease was focal. As is shown in Fig. 3, the increase in tuberculosis cases between 1984 and 1992 occurred mainly in the most populous states with New York leading the way. It is worth noting, however, that even among the states that had no increase between 1984 and 1992 the overall percentage decline was less than in previous years.

Perhaps the defining event in attracting public attention to the seriousness of the resurgence of tuberculosis was the death of a prison guard in upstate New York from multidrug-resistant tuberculosis. That this unfortunate occurrence took place against an emerging backdrop of increasing rates of tuberculosis caused by organisms resistant to at least isoniazid and rifampicin succeeded in alerting the public and policy-makers to the need for increased support for tuberculosis control at the national level (Frieden et al 1993). Hence, the Centers for Disease Control funding for tuberculosis increased from approximately US$45 million in 1992 to US$140 million in 1993.

In New York City, the epicentre of both the overall resurgence and the outbreaks of multidrug-resistant tuberculosis, funding for tuberculosis control increased from US$4 million in 1988 to a peak of more than US$40 million in 1994 (Frieden et al 1993, 1995, Fujiwara & Sherman 1997). In all, since the

Foreign-born cases secondary cases

Foreign-born cases secondary cases

US-born cases secondary cases

US-born cases secondary cases

FIG. 4. Interactions between US-born and foreign-born tuberculosis cases in San Francisco 1993-1994. Data from Chin et al (1998).

resurgence began, an estimated total of US$791 million has been spent in New York City to restore control of the disease (Fujiwara & Sherman 1997). Control of tuberculosis has been restored in New York City and in the US as a whole, but it has been an expensive lesson. To avoid repeating the past failures we will have to examine carefully the epidemiology of the disease and to define and focus strategies more precisely and efficiently. In San Francisco we have been using a combination of conventional and molecular epidemiology to examine the dynamics of tuberculosis (Small et al 1994). These investigations have enabled determination of the interactions among risk factors for tuberculosis and the behaviour of the epidemic over time.

0 0

Post a comment