|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
ABSTRACT |
|---|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
The New York City tuberculosis (TB) case rate declined from 1991 to 1994 following more than a decade of increases. The present study investigated TB trends in New York City neighborhoods and their association with neighborhood-specific rates of application of directly observed therapy (DOT). Using Poisson regression models, TB trends in each of New York City's 30 health districts were classified as increasing, decreasing, or stable, as indicated respectively by significant positive, negative, or nonsignificant regression coefficients. Case counts increased in four health districts, decreased in 10, and were stable in 16. Decreasing TB was associated with a higher rate of application of DOT. TB cases among foreign-born persons increased in 12 health districts and were stable in 18, whereas cases among persons born in the United States decreased in 19 and were stable in 11 districts. Among the foreign-born, increasing TB was not associated with a lesser rate of application of DOT. These data provide some support for the role of DOT in containing TB, but also suggest that the application of DOT among foreign-born residents is less effective than among United States-born residents. This may be due to a greater proportion of TB cases among the foreign-born being due to reactivation of TB rather than new infection.
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
After decades of nearly continuous decline, rates of newly reported cases of active tuberculosis (TB) in New York City began to rise in 1979, peaking at 46.7 per 100,000 in 1991 (1). The increase in the city's TB case rate has been attributed to immigration from TB-endemic countries (2, 3), the human immunodeficiency virus/acquired immune deficiency syndrome (HIV/ AIDS) epidemic (4, 5), homelessness (6), household crowding (7), intravenous drug use (8), and severe financial cutbacks in the New York City TB control program during the fiscal crisis of the 1970s (6, 9, 10).
More recently, though, TB case rates in New York City have been on the decline. Age-adjusted TB case rates in New York City fell by 12.4% between 1991 and 1994, to 40.9 cases per 100,000. Expansion of the directly observed therapy (DOT) program beginning in 1992 has been cited as a crucial factor in this decline (11). An examination of TB case numbers by age, sex, race, place of birth (United States born versus foreign born), and type of TB (pan susceptible versus multidrug resistant) revealed an overall citywide decrease in TB morbidity for all subanalyses save place of birth (11). Yet the picture is more complex than that conveyed by citywide averages, as we demonstrate in this report. We describe trends in TB case rates in New York City as a whole and among its constituent health districts, with the aims of: (1) describing the geographic diversity in TB rates and trends within New York City; (2) elucidating the role of DOT in the citywide decline in TB observed from 1991 to 1994; and (3) assessing the contribution of TB among the foreign-born residents of New York City to the TB epidemic and the evidence for the utility of DOT in controlling this portion of the TB epidemic.
| |
METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
The present analysis investigated the associations between TB case counts in New York City and calendar year. Cases of TB were dated according to their year of case verification, which in some cases differed from the year of first positive culture. Data used were those for the period 1991 to 1994 for New York City as whole and for each of its constituent health districts. These geographic subunits have been used by the New York City Department of Health since the 1930s to analyze geographic heterogeneity in disease statistics. For the year 1990, health-district population counts ranged from 105,661 to 502,102, with a median of 237,399. Thus, New York City's health districts are comparable in population size to medium-sized cities in the United States. Data were provided by the Bureau of Tuberculosis Control of the New York City Department of Health. Included were the age, race, place of birth (United States or non-United States) of each TB case, and whether or not DOT was provided.
Statistical analyses employed log-linear Poisson regression models,
allowing for over- and underdispersion where appropriate, through
the equation log(E(Y)) = 
0 + 1X, where Var(Y) = 
2E(Y) (12).
The dispersion parameter 2 was estimated as the Pearson chi-square
statistic divided by the degrees of freedom, using the SAS program
PROC GENMOD (SAS Institute, Inc., Cary, NC) (13). TB trends
were classified as changing or stable according to whether the estimate of 1 was statistically significant (p < 0.05) or not; those locations with a significant positive estimate were classified as having an
increasing trend, those with a significant negative estimate were said
to be having a decreasing trend, and those with a nonsignificant estimate were classified as stable.
A separate analysis of the years 1991 to 1994 was done for TB case numbers among the foreign-born and United States-born populations. Foreign-born persons consist of residents of New York City who were born outside the United States and its territories, excepting those persons born abroad to United States citizens.
The cumulative percentages of cases treated with DOT in each health district for the years 1991 to 1994 were calculated. The distributions of rates of DOT application among foreign-born and United States-born TB cases were then compared, using the Kolmogorov- Smirnov test (14). The relationship between the application of DOT and TB trends was assessed by: (1) comparing the percentage of TB cases treated with DOT among the different types of health districts, using the Kruskal-Wallis test; and (2) regressing the health-district-specific trends (as summarized by the coefficients from the Poisson regression models) against the health-district-specific percentages of TB cases treated with DOT. This last regression was done with weighted least squares, with the weights equal to the inverse variances of the Poisson regression model coefficients, in order to incorporate the precision of the trends into the test of association.
| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
New York City reported 678 fewer cases of tuberculosis in
1994 than in 1991. Three health districts, all located in Manhattan, the city's central borough and representing only 8.9%
of the city's population, accounted for 52.1% of this decline
(Central Harlem, 124 cases; the Lower East Side, 141 cases;
the Lower West Side, 88 cases). The relative change in the expected value of the number of citywide TB cases was 
7.8%
per year (p < 0.01). At the level of health districts, however,
trends varied: TB cases decreased in 10 health districts, increased in four, and were stable in 16.
Figure 1A and B displays the number of TB cases by year for health districts experiencing upward or downward trends. The percentage of New York City's population residing in districts of decreasing trend was 26.7% (2.0 million persons); 18.0% (1.3 million persons) lived in districts where TB was increasing and 55.3% (4.1 million persons) lived in districts where TB was stable. Health districts with decreasing TB case counts from 1991 to 1994 had a higher incidence of TB cases than did health districts with increasing trends: the mean number of TB cases over the period from 1991 to 1994, weighted by the size of each health district's population, was 157.5 in the decreasing-trend districts; 102.4 in stable districts; and 74.0 in increasing-trend districts.
|
The percentage of TB cases treated with DOT in New
York City increased from 7% in 1991 to 20% in 1992, 34% in
1993, and 47% in 1994. Averaged across the period 1991 to
1994, 26% of all TB cases were treated with DOT. By health
district, the percentage of cases treated with DOT ranged from
12.9% in Maspeth, Queens, to 40.0% in Morrisania, Bronx. The
median percentage of cases treated with DOT was 27.8% among
health districts with decreasing TB, 21.5% among those with
stable TB, and 17.1% in the four health districts in which TB
was determined to be on the increase, but the difference was
not statistically significant (Kruskal-Wallis chi-square = 5.7, p = 0.06). Regressing the estimates of 1 against the percentage of TB cases treated with DOT (using weighted least-squares regression), however, resulted in a significant regression coefficient (p < 0.0001). The square of the sample correlation coefficient, r2, was 0.63.
Subanalysis of Tuberculosis by Place of Birth
During the years 1991 to 1994, 30% of New York City's TB cases occurred in persons born abroad. By health district, the percentage of TB occurring among the foreign born ranged from 8% in Central Harlem (Manhattan) to 61% in Corona (Queens), with a medium of 30%.
The relative change in the expected number of TB cases among the foreign-born population of New York City was 10.5% per year (p < 0.001). There were 12 health districts classified as having increasing TB case counts during the years 1991 to 1994; 17 districts were deemed to be stable and one had decreasing case counts, by our analysis. Thus TB in the foreign-born was largely unchanged or increasing from 1991 to 1994. Trends in TB case counts among foreign-born New York City residents for the 29 health districts with nondecreasing numbers of TB cases are depicted in Figure 2A and B. Of the 17 health districts in which TB among foreign-born New York City residents was classified as stable during the years 1991 to 1994, case numbers reached peaks in two health districts in 1991, in three districts in 1992, in three districts in 1993, and in nine districts in 1994.
|
The relative change in the expected number of TB cases
among the United States-born population of New York City
was 14.7% per year (p < 0.01). At the level of health districts, 19 districts had decreasing case numbers and 11 were
stable. Thus, TB among persons born in the United States was
decreasing or unchanged from 1991 to 1994. Trends in TB
among United States-born New York City residents are depicted in Figure 2C and D. Of the 11 health districts in which
TB among United States-born New York City residents was
classified as stable during the years 1991 to 1994, TB peaked in
10 health districts in 1992. Analysis of temporal trends in these
health districts could have yielded different results if the time
interval studied had been shifted up by one year.
During the years 1991 to 1994, United States- and foreign-born cases of TB had similar distributions with regard to the
percentage of cases treated with DOT (Kolmogorov-Smirnov
test statistic = 0.77, p = 0.59). For United States-born cases,
the percentages ranged from 12.5% to 39.6% (median = 23.8%); for foreign-born cases, the percentages ranged from
11.3% to 44.0% (median = 23.9%). With respect to TB cases
among foreign-born New York City residents, the median percentage of TB cases treated with DOT was 27.1% among health districts with stable TB, 20.5% among those with increasing TB, and 31.9% in the one health district with declining TB. Omitting the one health district in which foreign-born
TB was declining, the difference in the application of DOT
was not significant (Kruskal-Wallis chi-square = 1.5, p = 0.22).
This lack of significance persisted even when districts with
nonincreasing case numbers were divided into two groups according to whether their estimates of 1 were positive or negative (Kruskal-Wallis chi-square = 2.3, p = 0.31). Regressing
the estimates of 1 against the percentage of TB cases treated
with DOT (using weighted least-squares regression), the regression parameter was nonsignificant (p = 0.15) and r2 was
0.07. Figure 3A and B shows trends and DOT application
rates for TB among foreign-born New York City residents.
|
With respect to TB cases among United States-born New
York City residents, the median percentage of TB cases treated
with DOT was 18.8% among health districts with stable TB
and 27.4% among those with decreasing TB. The difference in
the application of DOT was highly significant (Kruskal-Wallis
chi-square = 7.0, p < 0.01). When we regressed the estimates
of 1 against the percentage of TB cases treated with DOT
(using weighted least-squares regression), the regression coefficient was significant (p < 0.0001) and r2 was 0.46. Figure 3C
and D shows trends and DOT application rates for TB in
United States-born New York City residents.
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Our analysis illustrates how the use of citywide summary statistics to describe a TB epidemic in a city as large and diverse as New York City can obscure geographic variability in TB trends. Although mixing of persons from different health districts and different places of birth certainly occurs (and has consequences for TB transmission), our analysis at the level of health districts and by place of birth reveals meaningful differences in TB patterns. Our study shows that locally, TB was not declining in areas where a majority of New York City's population resided during the years 1991 to 1994. Moreover, increases in TB among foreign-born persons were present in more than one-third (12 of 30 districts) of New York City's health districts.
We found evidence for a general association between DOT prevalence and changes in TB incidence. We determined that 63% of the variation in TB trends overall and 46% of the variation in TB trends among United States-born persons (as determined from the coefficients of the Poisson model) could be explained by the percentage of TB cases that had been treated with DOT. As the DOT program expanded during the years 1991 to 1994, TB rates declined 12.4% in New York City, whereas in Baltimore, TB rates declined 31.5% during the first 4 yr of the implementation of community-based DOT in that city (15). However, Baltimore differs from New York City in at least two substantial ways. First, only 3.9% of all TB cases in Baltimore during the years 1987 to 1992 occurred among foreign-born residents of Baltimore, a situation that contrasts sharply with the 30% figure for New York City. Second, in 1990, Baltimore was a city of only approximately 726,000 persons, where New York City's population was 7.3 million. There may be difficulties in producing similar results within a city like New York, with more than 10 times the population and nearly four times the physical size (16).
The number of TB cases among foreign-born New York
City residents has been on the rise, and the increase is too large
to be explained away by a concomitant increase in the foreign-born population of New York City. (Estimated changes in TB
rates among the foreign born by borough, obtained by interpolating between 1990 U.S. Census data and the 1995 Intercensal projection, were 40.1%, 26.6%, 12.9%, 31.1%, and 195.4%,
in the Bronx, Brooklyn, Manhattan, Queens, and Staten Island,
respectively [17].) The most likely explanation for this result is
that more recently arrived foreign-born residents have a
higher infection rate or a higher rate of active disease than do
longer-term foreign-born residents. Because data on length of
stay in the United States were not available for foreign-born TB cases, it is not possible to draw a definite conclusion about the reasons for increasing TB among the foreign-born in New
York City. However, McKenna and colleagues have found
that TB among immigrants generally occurs within the first 5 yr
of their arrival (2).
In New York City, there appears to be a lack of association between TB trends among the foreign born and the relative number of foreign-born cases treated with DOT. This suggests that the treatment of active cases of TB with DOT, although assuring compliance with medical therapy (18) and thus reducing the incidence of secondary cases, has not halted that portion of the TB epidemic that originates in distant countries. It is possible that a higher rate of use of DOT, such as at the level applied in Baltimore (where over 50% of city-managed TB cases have been treated annually with DOT since 1984 [15]), might be necessary to produce a decline in TB case rates among the foreign-born population of New York City, or at least stabilize these rates. In addition, TB-control efforts, such as routine skin-test screening followed by preventive therapy, might make a difference in health districts with high foreign-born case rates.
Wang and colleagues, in their study of TB among Asian immigrants to British Columbia, Canada, concluded that undiagnosed active TB at the time of immigration was an important factor in the increased incidence of TB among recent immigrants (19). If this holds true for New York City as well, then a more careful medical inspection, followed by the initiation of TB treatment at the time of immigration, may be needed in order to reduce the incidence of TB among the city's foreign-born residents.
Several limitations of our analysis should be borne in mind. TB case rates are a function of incidence, reporting, and diagnosis, all of which can vary from year to year. In our study, TB cases were dated using the year of their verification rather than the year of first positive culture, which was not available for all reported cases. When year of first positive culture was available, the difference between it and the year of verification varied by health district (data not shown), possibly affecting our results. Increased diagnosis of TB in the wake of climbing TB rates during the 1980s is undoubtedly another phenomenon whose level varies from one health district to another. Unfortunately, no data were available that would have allowed us to correct for this problem. In addition, our analysis is based on TB case numbers rather than on age-adjusted TB case rates by place of birth. The results of this analysis therefore might be somewhat sensitive to population changes at the health-district level, although this seems unlikely, since estimates of population changes during the study period are not commensurate with the observed changes in TB incidence.
With respect to DOT, it is possible that an ecologic fallacy explains some or all of our findings. Our finding that numbers of TB cases decreased most rapidly in areas with the highest levels of application of DOT does not establish cause and effect. In the pre-AIDS era, even without a broad-based program of DOT, TB did decline in New York City. In the 1990s, DOT may have been implemented most vigorously in areas that were ready to experience a decline in TB case rates as a result of factors other than DOT. Declines, where they have occurred, may be due in part to what Blower and associates describe as the "natural behavior of an epidemic" (20) (i.e., the tendency for TB epidemics to rise and fall even in the absence of treatment). It is possible that declining TB among United States-born persons in areas of high incidence may result in part from an exhaustion of the supply of persons highly susceptible to active TB (i.e., the HIV-infected). Because HIV status is unavailable for the vast majority of TB cases, it is not possible to easily sort out the effect of the HIV/AIDS epidemic on the TB epidemic. Recent declines in HIV-associated mortality in New York City (21) might have been accompanied by changes in the incidence of opportunistic infections, including TB. Further research needs to be done to better understand the confluence of these two epidemics.
Additionally, the analysis reported here focuses on a relatively short time period, specifically the 4 yr from 1991 to 1994. There are inevitable statistical problems in analyzing such a small span of years. An analysis based on additional years of data might have shown that TB among United States-born persons was on the decline in a greater proportion or perhaps all of the health districts of New York City.
Although acknowledging these sources of uncertainty, we conclude that the decline in the citywide TB case rate in New York City from 1991 to 1994 was driven by the contributions of several health districts that began with high rates. The increased application of DOT in 1992 does not sufficiently explain why TB declined in some populations in some parts of the city and increased in others. An increasing TB rate among foreign-born New York City residents cannot be controlled only through the efforts of the New York City Department of Health. TB control programs need to be designed at both the national and local level so that declining TB is the norm for all New York City residents, both foreign born and United States born.
| |
Footnotes |
|---|
Correspondence and requests for reprints should be addressed to Dr. Amy L. Davidow, Department of Environmental Medicine, New York University School of Medicine, 341 East 25th Street, New York, NY 10010-2598.
(Received in original form December 16, 1996 and in revised form June 16, 1997).
Dr. Davidow is an Aaron Diamond Foundation Fellow.Acknowledgments: The writers thank the Bureau of Tuberculosis Control of the New York City Department of Health for the provision of TB case records; and William N. Rom, M.D., Chief, and Neil Schluger, M.D., Division of Pulmonary and Critical Care Medicine of the Bellevue Hospital Center for many helpful conversations.
Supported by Grant HL51517 from the National Heart Lung and Blood Institute; Grant DA06001 from the National Institute on Drug Abuse; Center for AIDS Research Grant 1P30AI27742 from the National Institute on Allergy and Infectious Diseases; a grant from the Aaron Diamond Foundation; and funds from the Consolidated Edison Company of New York.
| |
References |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
1. Bureau of Tuberculosis Control. 1992. Tuberculosis in New York City, 1991: Information Summary. New York City Department of Health, New York.
2.
McKenna, M. T.,
E. McCray, and
I. Onorato.
1995.
The epidemiology of
tuberculosis among foreign-born persons in the United States, 1986 to
1993.
N. Engl. J. Med.
332:
1071-1076
3. Centers for Disease Control. 1990. Tuberculosis among foreign-born persons entering the United States: recommendations of the Advisory Committee for the Elimination of Tuberculosis. Morbid. Mortal. Weekly Rep. 39(RR-18):1-21.
4. Barnes, P. F., A. B. Bloch, P. T. Davidson, and D. E. Snider Jr.. 1991. Tuberculosis in patients with human immunodeficiency virus infection. N. Engl. J. Med. 324: 1644-1650 [Medline].
5.
Rieder, H. L.,
G. M. Cauthen,
G. W. Comstock, and
D. E. Snider Jr..
1989.
Epidemiology of tuberculosis in the United States.
Epidemiol. Rev.
11:
79-98
6. Brudney, K., and J. Dobkin. 1991. Resurgent tuberculosis in New York City: human immunodeficiency virus, homelessness, and the decline of the tuberculosis control programs. Am. Rev. Respir. Dis. 144: 745-749 [Medline].
7. Drucker, E., P. Alcabes, W. Bosworth, and B. Sckell. 1994. Childhood tuberculosis in the Bronx. Lancet 343: 1482-1485 [Medline].
8.
Friedman, L. N.,
M. T. Williams,
T. P. Singh, and
T. R. Frieden.
1996.
Tuberculosis, AIDS, and death among substance abusers on welfare
in New York City.
N. Engl. J. Med.
334:
828-833
9. Council of Lung Associations of New York. 1981. New approach to a resurging crisis. Report of the Task Force on Tuberculosis in New York, 1980. Tuberculosis and Respiratory Disease Association of New York, New York.
10. New York Lung Association. 1979. Tuberculosis in New York City 1978. Tuberculosis and Respiratory Disease Association of New York, New York.
11.
Frieden, T. R.,
P. I. Fujiwara,
R. M. Washko, and
M. A. Hamburg.
1995.
Tuberculosis in New York City
turning the tide.
N. Engl. J. Med.
333:
229-233
12. McCullagh, P., and J. A. Nelder. 1989. Generalized Linear Models. Chapman & Hall, London. 193.
13. SAS Release 6.11 SAS Institute, Inc., Cary, NC.
14. Siegel, S., and N. J. Castellan, Jr. 1988. Nonparametric Statistics for the Behavioral Sciences. McGraw-Hill, New York. 144.
15.
Chaulk, C. P.,
K. Moore-Rice,
R. Rizzo, and
R. E. Chaisson.
1995.
Eleven
years of community-based directly observed therapy for tuberculosis.
J.A.M.A.
274:
945-951
16. U.S. Bureau of the Census. 1994. County and City Data Book: 1994. U.S. Government Printing Office, Washington, DC.
17. Population Division. 1996. Population Estimates for Counties (sorted alphabetically within state): July 1, 1990 to July 1, 1995. U.S. Bureau of the Census.
18. Reichman, L. B.. 1996. How to ensure the continued resurgence of tuberculosis. Lancet 347: 175-177 [Medline].
19. Wang, J. S., E. A. Allen, D. A. Enarson, and S. Grzybowski. 1991. Tuberculosis in recent Asian immigrants to British Columbia, Canada: 1982-1985. Tubercle 72: 277-283 [Medline].
20. Blower, S. M., A. R. McClean, T. C. Porco, P. M. Small, P. C. Hopewell, M. A. Sanchez, and A. R. Moss. 1995. The intrinsic transmission dynamics of tuberculosis epidemics. Nature Medicine 1: 815-821 [Medline].
21. Chiasson, M., L. Berenson, W. Li, S. Schwartz, B. Mojica, and M. Hamburg. 1997. Declining AIDS mortality in New York City. In Program and Abstract of the Fourth Conference on Retroviruses and Opportunistic Infections, 1997. Infectious Diseases Society of America for Retrovirology and Human Health, Alexandria, VA.
This article has been cited by other articles:
 |
|
 |
|
  F. M. Gany, C. Trinh-Shevrin, and J. Changrani Drive-by Readings: A Creative Strategy for Tuberculosis Control Among Immigrants Am J Public Health, January 1, 2005; 95(1): 117 - 119. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Davidow, B. T. Mangura, E. C. Napolitano, and L. B. Reichman Rethinking the Socioeconomics and Geography of Tuberculosis Among Foreign-Born Residents of New Jersey, 1994-1999 Am J Public Health, June 1, 2003; 93(6): 1007 - 1012. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. MATTEELLI, C. CASALINI, M. C. RAVIGLIONE, I. EL-HAMAD, C. SCOLARI, E. BOMBANA, M. BUGIANI, M. CAPUTO, C. SCARCELLA, and G. CAROSI Supervised Preventive Therapy for Latent Tuberculosis Infection in Illegal Immigrants in Italy Am. J. Respir. Crit. Care Med., November 1, 2000; 162(5): 1653 - 1655. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Proc. Am. Thorac. Soc. | Am. J. Respir. Cell Mol. Biol. |