Decrease in Reported Tuberculosis Cases—United States, 2009 FREE

MMWR. 2010;59:289-294

1 figure, 1 table omitted

Every year, CDC reports results from the National TB Surveillance System for the previous year. For 2009, a total of 11,540 tuberculosis (TB) cases were reported in the United States. The TB rate was 3.8 cases per 100,000 population, a decrease of 11.4% from the rate of 4.2 per 100,000 reported for 2008. The 2009 rate showed the greatest single-year decrease ever recorded and was the lowest recorded rate since national TB surveillance began in 1953.¹ TB case counts and rates decreased substantially among both foreign-born and U.S.-born persons, although foreign-born persons and racial/ethnic minorities continued to have TB disease disproportionate to their respective populations. The TB rate in foreign-born persons was nearly 11 times higher than in U.S.-born persons. The rates among Hispanics and blacks were approximately eight times higher than among non-Hispanic whites, and rates among Asians were nearly 26 times higher. The large decrease in reported cases during 2009 might represent a decrease in TB disease resulting from changes in population demographics or improved TB control. However, increased underreporting or underdiagnosis of TB also is possible. CDC currently is investigating possible causes for the sharp decrease in reported TB cases. Diagnosing and reporting all TB cases is essential to ensure treatment of patients with TB and implementation of other public health interventions that interrupt transmission.

Health departments in the 50 states and the District of Columbia (DC) electronically report to CDC the TB cases that meet the CDC/Council of State and Territorial Epidemiologists case definition.* Reports include the patient's race, ethnicity (i.e., Hispanic or non-Hispanic), treatment information, and, whenever available, drug-susceptibility test results. CDC calculates national and state TB rates overall and by racial/ethnic population using current U.S. Census population estimates. Annual estimates were used to calculate the national TB rate and the percentage change from 2008 to 2009. Population denominators used to calculate TB rates and percentage changes over time according to national origin (U.S.-born versus foreign-born persons) were obtained from the Current Population Survey. A U.S.-born person was defined as someone born in the United States or in its associated jurisdictions, or someone born in a foreign country but having at least one U.S.-born parent. Persons not meeting this definition were classified as foreign born. For 2009, patients with unknown country of birth represented 2.0% (235 of 11,540) of total cases. For this report, persons identified as white, black, Asian, American Indian/Alaska Native, native Hawaiian or other Pacific Islander, or of multiple races were all classified as non-Hispanic. Persons identified as Hispanic might be of any race.

In 2009, TB rates in the 51 reporting areas ranged from 0.4 (Wyoming) to 9.1 (Hawaii) cases per 100,000 population (median: 2.7 cases per 100,000 population). Thirty-six states and DC had lower rates in 2009 than 2008; 14 states had higher rates. Four states (California, Florida, New York, and Texas) reported more than 500 cases each for 2009. Combined, these four states accounted for half (50.3% [5,801]) of all TB cases in 2009.

In 2009, a total of 4,499 TB cases were reported in U.S.-born persons (representing 39.8% of the 11,305 cases with known national origin), compared with 5,282 reported in 2008. The 2009 rate in U.S.-born persons was 1.7 per 100,000, a decrease of 15.8% compared with 2008, and a decrease of 77% compared with 1993 (Figure). In 2009, a total of 1,861 cases were reported among blacks, representing the highest number of TB cases among U.S.-born persons and 41.4% of all U.S.-born cases in 2009. A total of 3,386 cases were reported among Hispanics, more than any other racial/ethnic group, followed by Asians and blacks. Asians had the highest TB case rate among all racial/ethnic groups. From 2008 to 2009, TB rates decreased for all racial/ethnic minorities. The greatest annual decrease in TB rate was among whites (15.2%), followed by blacks (14.0%) and Hispanics (13.6%). The smallest decrease in 2009 was among Asians (9.0%).

A total of 6,806 TB cases were reported in foreign-born persons in 2009, compared with 7,602 reported in 2008, a decrease of 10.5%. For 2009, these cases represented 60.2% of all cases with known national origin, compared to 59.0% of cases with known origin in 2008. The TB rate among foreign-born persons in 2009 was 18.6 per 100,000 population, a 9.0% decrease compared to 2008 and a 45.3% decrease since 1993. In 2009, four countries accounted for 50.1% of TB cases in foreign-born persons: Mexico (1,574), the Philippines (799), India (523), and Vietnam (514).

In 2009, among persons with TB with a known human immunodeficiency virus (HIV) test result, 10.2% (690 of 6,743) were coinfected with HIV. California and Vermont data were not available for this calculation.†

TB cases are classified by site of disease (pulmonary or extrapulmonary) and whether Mycobacterium tuberculosis was cultured from a patient specimen (i.e., culture positive or culture negative). In 2009, a total of 8,535 TB cases were pulmonary TB, of which 7,133 (83.6%) were culture positive. Of 2,297 extrapulmonary TB cases, 1,630 (71.0%) were culture positive. Site of disease or culture status was unknown for 708 cases. From 2008 to 2009, culture-positive pulmonary cases decreased 13.6% (8,257 to 7,133), culture-negative pulmonary TB cases decreased 17.5% (1,700 to 1,402), culture-positive extrapulmonary TB cases decreased 8.3% (1,777 to 1,630), culture-negative extrapulmonary TB cases decreased 3.1% (688 to 667), and cases with unknown site of disease or culture status increased 48.7% (476 to 708).

Drug-susceptibility data generally are available 1 year later than provisional surveillance data. The most recent year for which complete drug-susceptibility data are available is 2008. Drug-susceptibility test results for isoniazid and rifampin were reported for 96.0% (9,628 of 10,034) of culture-confirmed TB cases in 2008, compared with 97.9% (10,251 of 10,468) in 2007. A total of 107 cases of multidrug-resistant TB (MDR TB)‡ were reported in 2008, which represented 1.1% of all culture-positive cases with drug-susceptibility testing, compared with 1.2% [123 of 10,251 cases] in 2007. For persons with a previous history of TB, the percentage of cases with MDR TB remained approximately four times higher than for persons without a prior history of TB. In 2008, foreign-born persons accounted for 77.6% of MDR TB cases. Foreign-born persons had higher percentages of MDR TB than U.S.-born persons, both among persons with (4.3% versus 1.4%) and without (1.0% versus 0.4%) a previous history of TB. To date, no new cases of extensively drug-resistant TB (XDR TB)§ have been reported in 2009.

The recommended length of drug therapy for most types of TB is 6-9 months. In 2006, the latest year for which end-of-treatment data are complete, 83.7% of patients for whom <1 year of treatment was indicated completed therapy within 1 year, compared with 83.1% in 2005.

C Winston, R Pratt, L Armstrong, T Navin, Div of TB Elimination, National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention, CDC.

The 11.4% decrease in reported TB rate in 2009 is the largest single-year decrease ever recorded. From 1953 to 1993, the single largest annual percentage decrease in TB case rate was 11.1% in 1956.¹ During 1993-2000, TB case rates decreased an average of 7.8% each year (range: 4.8%-8.5%), and for 2000-2008, the rate decreased an average of 3.8% annually (range: 2.3%-6.7%).^1,3 The decrease for 2009 is unusual and unexpectedly large. CDC and the National Tuberculosis Controllers Association (NTCA) are investigating whether the decrease represents a reduction in disease rate resulting from improved TB control or population demographic shifts, or might be the result of underdiagnosis or underreporting of disease.

In 2009, the national tuberculosis case definition was modified for the first time since 1996.⁴ For the first time, the updated case verification algorithm incorporated nucleic acid amplification tests,⁵ interferon gamma release assays,⁶ and the use of computerized tomography scans of the chest. However, because cases verified using these diagnostic tests were eligible previously for reporting as “provider-diagnosed” cases, the 2009 revised case definition should not have caused a decrease in case reports. Also, these changes do not have any effect on the reporting of culture-positive TB cases, which decreased 12.7% from 2008 to 2009. CDC and NTCA are working with state and local health departments to ensure that all laboratory and clinically confirmed TB cases were reported accurately in 2009.

In 2009, case definition changes described above were incorporated into the national Report of Verified Case of Tuberculosis (RVCT), the standardized data-collection instrument used by health departments to report data to CDC.⁴ CDC and NTCA will examine whether implementation of the revised RVCT was associated with longer than expected delays in reporting for 2009.

In addition to changes to the case definition and RVCT, major changes to software systems used to collect and report cases electronically also occurred in 2009. Many states experienced difficulties transmitting data; however, CDC has verified that TB case counts received via electronic reporting for 2009 were correct. In addition, software systems are an unlikely cause of the 2009 decline because declines in TB case reports for 2009 occurred across all types of state software systems and across all tiers of TB case burden, including low, medium, and high incidence states.

Part of the 2009 decrease in foreign-born cases also might be attributed to new technical instructions issued in 2007⁷ for preimmigration TB screening. These instructions might have prompted more TB diagnoses and treatment before immigration because, for the first time, they require culture of respiratory specimens in immigrants and refugees who are suspected of having pulmonary TB based on chest radiographs. The previous technical instructions only required smear microscopy, which is less sensitive and has been shown to result in fewer TB diagnoses. Cases of TB in patients who are diagnosed and begin treatment before immigration are not counted in the U.S. surveillance system because they are not incident TB cases in the United States. CDC plans to compare TB case reports among foreign-born persons by length of time in the United States and country of origin and conduct pre/post analyses based on implementation of the technical instructions. The 2009 decrease might also have resulted, in part, from reductions in immigration and increases in recent immigrants returning to their native country, especially Mexico. Some anecdotal reports from state and local health departments indicate that they have seen fewer TB patients who are recent immigrants.

The findings in this report are subject to at least two limitations. First, the analysis was based on provisional 2009 data that are subject to change. This applies to TB case counts and HIV data, both of which are incomplete at the time of this report. Additional data could influence the results. Second, population denominator data are drawn from multiple U.S. Census sources and are subject to periodic adjustment in the estimates. CDC's annual TB surveillance summary, due to be published in fall 2010, will provide final data reflecting 2009 surveillance.

Decreases in TB case rates might reflect actual decreases in TB incidence. The reduction in culture-positive cases is slightly larger than the overall case decrease. Because a positive culture is the most definitive criterion for confirming a case of TB, reporting artifact related to changes in case definition or the clinical judgments of physicians is not a likely cause of the decrease in TB cases. If underreporting and underdiagnosis of TB cases can be excluded, determining the causes of such a large actual decrease in TB cases will be important, as will understanding which specific public health interventions related to the three U.S. priorities of (1) diagnosing and treating patients with TB disease, (2) conducting contact investigations of TB cases, and (3) targeted testing and treatment of latent TB infection are having the greatest impact, so these interventions can be reinforced and replicated.

The findings in this report are based, in part, on data contributed by state and local TB control officials.

The U.S. national tuberculosis (TB) case rate has been decreasing at approximately 4% each year; foreign-born persons and racial/ethnic minorities have represented a disproportionate number of reported TB cases.

What does this report add?

In 2009, preliminary data show an unexpectedly large decrease in the national TB case rate (11.4%), which might be the result of surveillance reporting changes, improved TB control efforts, changes in population demographics, or underdiagnosing or underreporting of disease.

What are the implications for public health practice?

Because undiagnosed or unreported cases can result in preventable spread of TB, all suspected cases of TB should be promptly evaluated, and all diagnosed cases should be promptly reported to public health authorities.

*Available at http://www.cdc.gov/ncphi/disss/nndss/casedef/tuberculosis_current.htm.

†For this report, data from Vermont and California were excluded from the analysis. Vermont no longer reports HIV status to CDC, and California has not reported since 2004.

‡Defined as a case of TB in a person with a Mycobacterium tuberculosis isolate resistant to at least isoniazid and rifampin.²

§Defined as a case of TB in a person with an M. tuberculosis isolate with resistance to at least isoniazid and rifampin among first-line anti-TB drugs, resistance to any fluoroquinolone (e.g., ciprofloxacin or ofloxacin), and resistance to at least one second-line injectable drug (e.g., amikacin, capreomycin, or kanamycin).²