Page 234
APPENDIX E
Estimating the Number of Tuberculosis Cases That Can Be Prevented by a Program of Screening and Preventive Therapy of Newly Arrived Immigrants to the United States from Countries with a High Rate of Tuberculosis and the Costs of the Program
The first step in estimating the number of cases that can be prevented by a program of screening and preventive therapy of newly arrived immigrants to the United States from countries with a high rate of tuberculosis was to define the high-risk countries. A recent report on the global burden of tuberculosis included appendixes with a variety of epidemiological measures of tuberculosis burden (Dye et al., 1999). One of the measures was the prevalence of infection for each country. The median prevalence of infection was 36 percent, so any country with a prevalence of infection of >35 percent was defined as a high-tuberculosis-risk country. Fifty-three high-tuberculosis-risk countries were identified, and a list of these countries and their estimated infection prevalences are included in Table E-1 .
Mexico had a prevalence of infection estimated at only 17 percent. Although this does not qualify as a high tuberculosis burden on the global scale, it is nearly 2.5 times the rate of infection estimated for the United States. Because of the large number of immigrants to the United States from Mexico, nearly one-quarter of all cases of tuberculosis among the foreign-born in the United States occur among individuals born in Mexico. Therefore, an argument can be made for the inclusion of immigrants from
Page 235
Mexico in a screening program to identify individuals with latent tuberculosis infection.
The 1996 Statistical Report of the Immigration and Naturalization Service shows that in 1996 421,405 immigrants were admitted to the United States as new arrivals. A total of 187,079 immigrants were admitted from 24 of the high-tuberculosis-risk countries. No immigrants were reported separately in the 1996 statistical report as arriving from the other 29 high-tuberculosis-risk countries, but 4,360 newly arrived immigrants were reported as arriving from other countries in the African, American, Eastern Mediterranean, Southeast Asian, and Western Pacific regions. However, these individuals are not included in further analyses since the precise country of origin could not be ascertained. An additional 52,946 newly arrived immigrants from Mexico were included in the analysis, yielding a total of 240,025 newly arrived immigrants from high-tuberculosis-risk countries and Mexico.
ESTIMATING THE NUMBER OF TUBERCULOSIS CASES
After estimating the number of immigrants to be included in a screening program, the second step is to estimate the number of tuberculosis cases that will occur among these immigrants. Two methods were used to estimate the number of tuberculosis cases. One uses the risk of developing tuberculosis given a positive tuberculin skin test in a population at high risk for infection and with a mixture of old and new infections. The other method uses country- and region-specific annual incidence rates estimated for newly arrived immigrants during the first 5 years in the United States.
The risk of tuberculosis given a positive tuberculin skin test in a population at high risk for infection and with a mixture of old and new infections can be calculated from data for the placebo group of the U.S. Public Health Service trial of isoniazid preventive therapy among household contacts. Patients were enrolled in this trial from 1956 to 1959. Even though nearly 90 percent of these individuals were enrolled as household contacts of individuals with new active cases of tuberculosis, this was very early in the era of tuberculosis chemotherapy (isoniazid was first discovered in 1952) and it is likely that this population had a mixture of recent and old tuberculosis infections. There were 121 cases of tuberculosis among the 4,992 individuals with a tuberculin skin test induration of 10 millimeters or more, for a 5-year risk of tuberculosis in the placebo group of 2.42 percent. By using the individual country estimates of the prevalence of infection and the number of newly arrived immigrants, a total of 87,287 individuals would be infected. With a risk of 2.42 percent, 2,112 cases would be expected to arise in this group in 5 years.
Page 236
Please refer to the page image for an unflawed representation of this content.
|
Page 237
Ivory Coast
0.36
WPRO
Cambodia
0.64
Kenya
0.36
People's Republic of China
0.36
Liberia
0.37
Hong Kong
0.37
Madagascar
0.37
Macau
0.48
Mauritania
0.36
Mongolia
0.45
Namibia
0.37
New Guinea
0.44
Nigeria
0.36
Philippines
0.47
South Africa
0.38
Republic of Korea
0.36
Swaziland
0.37
Vietnam
0.44
Togo
0.37
Zambia
0.36
Mexico
0.16
Zimbabwe
0.36
AMRO
Bolivia
0.47
Equador
0.38
Haiti
0.54
Peru
0.44
Page 238
Another method for estimation of the number of cases expected to arise from the group of immigrants mentioned above is to use the estimates of the risk of tuberculosis for immigrants during their first 5 years in the United States (McKenna et al., 1995) and estimates of region- and country-specific annual tuberculosis incidence rates for immigrants in the United States (Zuber et al., 1997). When estimates were available, estimates for individual countries were used; otherwise, the average incidence for immigrants from the region where that country is located was used. By this method, an average incidence of 173 cases of tuberculosis per 100,000 population is estimated for the immigrants from the high-risk countries and Mexico, producing a total of 2,081 cases per year among immigrants who have been in the United States for 5 years or less. A summary of the data obtained by both methods is provided in Table E-2 .
The estimates obtained by the two methods are remarkably similar, and this lends some confidence to the possibility that they may both be accurate means of obtaining estimates. Taking the average of the two estimates would yield an estimate of 2,097 cases among these newly arrived immigrants during their first 5 years in the United States. That also translates into 2,097 cases of tuberculosis per year among these immigrants when one includes all cases from all five annual cohorts (those in the United States for less than one year, those in the United States for 1 to 2 years, those in the United States for 2 to 3 years, etc.). By comparison of the estimates for immigrants from the seven countries that account for the majority of foreign-born tuberculosis cases in the United States, the estimates account for 27 percent of the foreign-born cases overall. This may be a reasonable proportion of the cases, since only about 55 percent of the cases of tuberculosis among newly arrived immigrants are estimated to occur in the first 5 years and cases will arise among undocumented individuals and other long-term residents who are not classified as new arrivals (e.g., those who have adjusted immigrant status, students, and workers).
Despite the similarity of the estimates obtained by the two methods, there are a number of weaknesses in the data and the calculation of these estimates. Immigrant flows into the United States are constantly changing, and the number of immigrants coming from any one country can change dramatically from year to year. The most recent data for the number of immigrants entering the United States was from a 1996 Immigration and Naturalization Service summary. A great deal may have changed in the intervening years. Similarly, even if the number of immigrants is the same, the characteristics of the immigrants may change over time. For example, the health status and general tuberculosis risk for immigrants arriving in the United States during a period of political instability in their home country may be very different from those immigrants arriving dur
Page 239
ing more stable times. The tuberculosis risks calculated by Zuber et al. (1997) and McKenna et al. (1995) used data from the late 1980s and early 1990s.
PREVENTION OF NEW CASES BY THE TREATMENT OF LATENT INFECTION
A 9- to 12-month course of isoniazid treatment is estimated to be about 75 percent effective in preventing tuberculosis in people with latent infection (this effectiveness measurement accounts for individuals who do not adhere to the prescribed regimen or who cannot complete the treatment because of drug intolerance). Rifampin-based regimens are also believed to have the same effectiveness as isoniazid-based regimens, and these might be used for immigrants from countries where isoniazid resistance is common. Therefore, effective treatment of 75 percent of the 2,097 potential tuberculosis cases would result in the prevention of 1573 cases of tuberculosis.
A rough estimate of the costs for a program that uses a 9-month regimen of self-administered isoniazid is provided in Table E-3 .
The cost estimates for mailings, calls, and visits are from a program that is used to monitor immigrants in Santa Clara County, California with Class A, B1, and B2 tuberculosis (Catlos et al., 1998) and the cost for medical services are current charges for tuberculosis services in metropolitan Denver, Colorado. The estimates assume that all individuals from high-tuberculosis-risk countries would be retested by the tuberculin skin test in the United States and that all tuberculin skin test-positive individuals would receive a chest radiograph and an evaluation by a physician. The costs for therapy are for 80 percent of those infected (assuming that 20 percent will have contraindications to therapy or will decline treatment), and will include the management of adverse reactions. The addition of directly observed therapy (DOT) would add about $100 million to the cost, assuming twice-weekly therapy and an additional cost of $20 per DOT visit. However, this would increase the effectiveness of the regimen and could lead to the prevention of 335 additional cases. Without DOT the cost per case prevented is slightly less than the cost of treating a case of tuberculosis, assuming a 60 percent hospitalization rate and including the costs of contact tracing, treatment of infected contacts for latent infection and treatment of secondary cases. However, there is no accounting for additional cases and the prevention of transmission of tuberculosis.
If immigrants are required to report for tuberculosis screening as a condition for receiving permanent residency cards (“green cards”), follow-up costs would be greatly reduced. If tuberculin skin test results were reliably reported from the preimmigration screening in the country
Page 240
of origin, that cost could be avoided. However, start-up costs for the program were not included in the calculations. Many public health departments will not be prepared for implementation of this kind of program and will incur significant costs in starting up programs for largescale tuberculin skin testing and treatment of individuals with latent infection. However, it must also be noted that these start-up costs will yield benefits to other programs for tuberculin skin testing and treatment of latent infection that will be needed in the same communities. Costs that would be incurred by the Immigration and Naturalization Service and the Division of Quarantine of the Centers for Disease Control and Prevention to ensure that information on the immigrants is reliably recorded and communicated between Immigration and Naturalization Service and the health department are not included. Costs are also not included for the training and monitoring of the individuals who will perform the tuberculin skin testing for the medical evaluations or for the additional contacts the immigrant that might be required to have with the Immigration and Naturalization Service after evaluation and therapy (if necessary) is completed and the permanent residency card is given out. However, the rapid and reliable flow of information and the training and monitoring should be a requirement for the current program, and many of these costs may be incurred whether screening requirements are changed or not.
Please refer to the page image for an unflawed representation of this content.
|
Page 241
Page 242
Please refer to the page image for an unflawed representation of this content.
|
Page 243
REFERENCES
Catlos et al. 1998 . Public health interventions to encourage TB class A/B1/B2 immigrants to present for TB screening . Am J Respir Crit Care Med 158(4) : 1037–1041 .
Dye, C. , Scheele, S. , Dolin, P. , Patania, V. , et al. 1999 . Global burden of tuberculosis: estimated incidence, prevalence, and mortality by country . JAMA 282 : 677–686 .
Immigration and Naturalization Service . 1996 . 1996 Statistical Report, Table 7, Immigrants admitted by type of admission and region and selected country of birth. Washington, DC: Immigration and Naturalization Service. 1995 .
McKenna M , McCray E , Onorato I. The epidemiology of tuberculosis among foreign-born persons in the United States, 1986 to 1993 . N Engl J Med 332 : 1071–1076 .
Zuber, P. , McKenna, M. , Binkin, N. , Onorato, I. , Castro, K. 1997 . Long-term risk of tuberculosis among foreign-born persons in the United States . JAMA 278 : 304–307 .