Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
1503 Infectious Disease Morbidity in the US Region Bordering Mexico, 1990–1998 Timothy J. Doyle1 and Ralph T. Bryan2,a 1 Division of Public Health Surveillance and Informatics, Epidemiology Program Office, and 2Division of Quarantine, National Center for Infectious Diseases, Centers for Disease Control and Prevention, Atlanta, Georgia The United States and Mexico share an international boundary ∼3000 km long. This border separates 2 nations with great differences in health status. The objective of this study was to assess morbidity due to infectious diseases in the US region bordering Mexico. The incidence between 1990 and 1998 of 22 nationally notifiable infectious diseases was compared between border and nonborder regions. Disease rates, reflected as rate ratios, were higher in the border region for botulism, brucellosis, diphtheria, hepatitis A, measles, mumps, rabies, rubella, salmonellosis, and shigellosis than in either of 2 nonborder comparison regions. These data indicate that incidence rates for a variety of infectious diseases of public health importance are significantly higher in the United States along the Mexican border than in nonborder regions. These results suggest that an inadequate public health infrastructure may contribute to excess morbidity due to infectious diseases in the border region. The United States and Mexico share an international boundary ∼3000 km long between 4 US and 6 Mexican states. Perhaps no other border in the world separates 2 nations with such discrepancies in health status, entitlements, and utilization [1]. The border region is commonly defined as the area in the United States and Mexico within 100 km of the border [2]. On the US side, the border region is characterized by lower than average socioeconomic status: 5 of the 14 poorest US counties are in the Texas borderlands [3]. In Mexico, by contrast, the northern border states are generally more prosperous relative to other states, and the border region economy is heavily influenced by commercial activity with the United States. In many respects, US and Mexican border communities have more in common with their sister communities across the international boundary than with communities in their own countries outside the border region. Thus, in these 2 distinct countries, there is a region of transition—i.e., the binational borderlands, which both separate and connect the United States and Mexico. During the past century, the binational border region has experienced steady population growth. Residents of border communities commonly live on 1 side and work on the other side of the border. Transborder human population movement is enhanced by American manufacturing plants located on the Mexican frontier. In one city alone, Ciudad Juarez, Chihuahua, Received 11 May 2000; revised 13 July 2000; electronically published 27 September 2000. Presented in part: International Conference on Emerging Infectious Diseases, Atlanta, March 1998. a Present affiliation: Office of the Associate Director for Minority Health, Centers for Disease Control and Prevention, Albuquerque, New Mexico. Reprints or correspondence: Mr. Timothy Doyle, CDC/EPO, Mail Stop K-74, 4770 Buford Hwy., Atlanta, GA 30341 ([email protected]). The Journal of Infectious Diseases 2000; 182:1503–10 q 2000 by the Infectious Diseases Society of America. All rights reserved. 0022-1899/2000/18205-0028$02.00 Mexico, 1200 companies have located plants, to take advantage of lower production costs [4]. This concentration of manufacturing activity has attracted residents to the border region from other parts of Mexico in search of employment opportunities. Legal northbound crossings of the border exceed 200 million annually, making this one of the world’s busiest international boundaries (unpublished data, US Immigration and Naturalization Service). Outside of urban areas, a number of US border residents live in unincorporated settlements referred to as “colonias.” An estimated 500,000 people live in 12500 colonias throughout the US border region, with most settlements in Texas [5]. Residents of colonias often lack access to indoor plumbing or adequate sewage and waste disposal systems [6]. Groundwater on both sides of the border has been shown to contain unhealthy concentrations of fecal coliform bacteria, thus posing a risk for waterborne infectious diseases [7, 8]. The border states monitor the health of their residents through standard, usually passive, public health surveillance methods. Such methods, however, fail to treat the population living in the border region as a single geographic or epidemiologic group, one that shares many of the same health and demographic features among states on both sides of the international boundary. The unique health characteristics of the borderland population have been qualitatively described, emphasizing the important contribution of infectious agents to morbidity and mortality [1]. In addition, quantitative studies have been conducted, addressing select diseases in specific border communities. Nevertheless, there have been few quantitative analyses of disease incidence along the border’s 3000-km length. Thus, the precise burden of infectious diseases in the region is unclear. The creation of a truly binational infectious disease surveillance system is made difficult by substantive differences in case definitions and laboratory disease confirmation methods be- 1504 Doyle and Bryan Figure 1. JID 2000;182 (November) Border and nonborder comparison regions, continental United States tween the United States and Mexico. Despite the close proximity and the many common characteristics of borderland communities, current disease reporting differences between the countries require separate enumeration of disease incidence for each country. We report here infectious disease morbidity, for a 9-year period, for the US region that borders Mexico. Methods Case report data for 1990–1998 were obtained via the National Notifiable Diseases Surveillance System (NNDSS) for 22 infectious diseases reportable to public health authorities in the United States [9]. The data included information on disease, the county and state reporting the case, the year of illness, and the age and ethnicity of the patient. Data on race were not consistently categorized throughout the study period and, therefore, were not analyzed extensively. The continental United States was divided into 3 regions (figure 1): region 1, 48 counties within ∼100 km of the border with Mexico; region 2, the remaining 312 counties in Arizona, California, New Mexico, and Texas 1100 km of the border; and, region 3, 44 nonborder states and the District of Columbia. Annual and 9-year mean, unadjusted, disease incidence rates were calculated for each region after exclusion of case reports from border states with missing county information. After we excluded other cases with missing age information, age-adjusted rates were computed by using the indirect standardization method [10]. Population estimates for 1994, the midpoint of the study period, were used as population denominators for all rate calculations (table 1) [11, 12]. Rate ratios were then calculated, which compared mean rates between region 1 and regions 2 and 3. Similar analyses were performed after we stratified for Hispanic ethnicity. During the 9-year study period, Table 1. regions. Demographic characteristics of border and nonborder Characteristic a Total population a Hispanic ethnicity, % b Age distribution, % 0–4 Years 5–17 Years 18–24 Years 25–44 Years 45–64 Years 164 Years a b Data from [11]. Data from [12]. Region 1 Region 2 Region 3 9,871,639 36 45,572,965 26 203,067,933 5 9 20 10 32 17 12 9 19 10 33 18 11 7 18 10 32 20 13 JID 2000;182 (November) US-Mexico Border Infectious Disease Morbidity 1505 Table 2. Total cases, average annual incidence, and rate ratios of incidence in border and nonborder regions for select nationally notifiable diseases, 1990–1998. Rate per 100,000 population (no. of cases) Disease Botulism, foodborne Brucellosis Cholera Diphtheria Haemophilus influenzae, c invasive disease Hansen’s disease Hepatitis A Hepatitis B Legionellosis Malaria Measles Meningococcal disease Mumps Pertussis Plague Rabies, human d Rubella e Salmonellosis e Shigellosis Tetanus Trichinosis Typhoid fever Region 1 Region 2 Region 3 Rate ratio a b Regions 1:2 Regions 1:3 0.03 0.18 0.02 !0.01 (31) (158) (14) (2) !0.01 (27) !0.01 (94) 0.09 (360) 0.03 (123) !0.01 (5) 0.02 (398) !0.01 (112) !0.01 (13) 5.30 2.03 0.53 1.85 6.78 8.17 2.57 3.16 0.35 0.10 29.89 6.54 0.29 0.45 4.54 0.88 0.97 2.05 !0.01 !0.01 0.15 18.94 17.81 0.02 !0.01 0.16 (339) (93) (26,552) (5811) (255) (403) (4031) (783) (863) (1817) (4) (3) (105) (5610) (5274) (21) (2) (145) 0.48 0.17 17.99 7.53 0.19 0.76 3.85 1.13 0.74 1.67 0.01 !0.01 0.09 17.49 14.22 0.02 !0.01 0.26 0.49 0.02 7.80 4.71 0.59 0.52 1.09 1.06 0.72 2.20 !0.01 !0.01 0.07 16.45 7.83 0.02 0.02 0.14 0.79 0.62 1.66 0.87 1.49 0.60 1.18 0.78 1.31 1.23 0.38 2.31 1.74 1.08 1.25 0.98 0.38 0.64 0.78 5.09 3.83 1.39 0.48 0.87 4.17 0.83 1.36 0.93 3.43 3.86 2.10 1.15 2.28 1.39 0.14 1.15 (1974) (694) (73,806) (30,874) (790) (3097) (15,792) (4639) (3049) (6837) (48) (6) (278) (23,918) (19,438) (99) (24) (1054) (8987) (376) (142,608) (86,021) (10,818) (9519) (19,903) (19,441) (13,069) (40,200) (24) (16) (1029) (100,220) (47,670) (310) (285) (2586) a Ratio of disease incidence in region 1 divided by disease incidence in region 2. Ratio of disease incidence in region 1 divided by disease incidence in region 3. Eight-year average annual incidence, 1991–1998 (excluding 1990). d Seven-year average annual incidence, 1992–1998 (excluding 1990 and 1991, when 875 cases in border states lacked information on county). e Three-year average annual incidence, 1996–1998 (excluding 1990–1995). b c temporal trends were examined by comparing annual incidence rates for each disease between the regions. The analyses were performed by using SAS statistical software (version 6.12 for Windows; SAS Institute, Cary, NC). Invasive Haemophilus influenzae disease was reportable in !80% of states during 1990, becoming nationally notifiable in 1991. Therefore, incidence for this disease was calculated for the 8-year period, 1991–1998. During 1990 and 1991, cases of rubella reported from California lacked information on county. Therefore, incidence of rubella in the 3 regions was calculated for the 7-year period, 1992–1998. Throughout the first 6 years of the study period, case reports and information on county were lacking from some states for cases of salmonellosis and shigellosis. Therefore, incidence rates for salmonellosis and shigellosis were calculated over 3 years, 1996–1998. Infectious disease morbidity in border and nonborder regions was further assessed by pooling incidence data from 2 groups of vaccine-preventable diseases. In both instances, pooling of such diseases was done throughout the 9-year period, after we excluded cases from border states with missing county information. Case reports of measles, mumps, and rubella were pooled, to assess incidence of diseases prevented by the conjugate measles-mumpsrubella (MMR) vaccine. Case reports of measles, mumps, rubella, diphtheria, pertussis, and tetanus (MMR-DPT) were also pooled, to assess the overall incidence of childhood vaccine-preventable diseases. Finally, case report data were not obtained for tuberculosis, AIDS, or other sexually transmitted diseases, because re- porting requirements differ for these diseases, and national data on these conditions are maintained on separate information systems. Results For the 9-year period, 1990–1998, we report the total cases of disease, the unadjusted mean annual disease incidence by region, and rate ratios between border and nonborder regions (table 2). Incidence rates in the border region (region 1) were greater than rates in nonborder counties of border states (region 2) for the following diseases: foodborne botulism, human rabies, brucellosis, diphtheria, rubella, hepatitis A, legionellosis, mumps, shigellosis, pertussis, measles, and salmonellosis (figure 2). In comparison with national disease rates (region 3), incidence rates in the border region were elevated for brucellosis, foodborne botulism, Hansen’s disease, measles, rabies, hepatitis A, plague, diphtheria, cholera, shigellosis, rubella, hepatitis B, tetanus, mumps, typhoid fever, and salmonellosis (figure 3). The same patterns were observed for ethnic-specific and age-adjusted disease incidence rates compared between regions (data not shown). When temporal trends were examined, the incidence of hepatitis A was consistently higher in the border region each year than in either of the 2 nonborder comparison regions. When 1506 Doyle and Bryan Figure 2. JID 2000;182 (November) Rate ratios of disease incidence in regions 1 and 2, 1990–1998 compared only with rates in region 3, the incidence of brucellosis and hepatitis B was also consistently higher in the border region during each year of the study period. During 1996–1998, when complete data were available, the incidence of salmonellosis and shigellosis was consistently higher in the border region than in either nonborder region. Further examination of temporal trends revealed that 24 of the 31 cases of foodborne botulism reported from the border region occurred in 1994. This observation suggested that a single common source outbreak was responsible for the high 9-year mean incidence in the border region. Indeed, most cases were part of an outbreak linked to a restaurant in El Paso, Texas [13]. When the analyses were stratified by ethnicity, the regional differences observed in the unstratified analyses generally persisted among both Hispanics and non-Hispanics. Although the incidence of hepatitis A in the border region was over twice as high among Hispanics than in non-Hispanics, non-Hispanics in the border region continued to show a similar magnitude of excess incidence of this disease, compared with non-Hispanics in nonborder regions. Similarly, for brucellosis, among both Hispanics and non-Hispanics, rates were higher in the border region than in nonborder regions, even though the overall rates were much higher among Hispanics. When age-adjusted analyses were performed that stratified by the age groups shown in table 1, no evidence of age confounding was observed. The population age distribution in the border region is slightly younger than that in nonborder regions. Although the mean age of hepatitis A cases in the border region (21 years) was younger than that in nonborder regions (23 and 28 years in regions 2 and region 3, respectively), excess incidence of hepatitis A in the border region persisted throughout all age groups. Similar results were observed for measles and shigellosis, with the highest rates observed throughout all regions in children !5 years old; yet greater incidence was observed in the border region within each age group. Overall, rate ratios derived from age-standardized incidence rates differed little from those based on unadjusted rates. Temporal trends for measles indicated a relative excess of this disease in the border region in 1990, 1992, and 1998. During 1990, 13300 measles cases occurred in the border region, representing a significant portion (12%) of the 127,000 cases reported nationally. For 1991, 2848 measles cases were reported throughout the 4 border states, yet 70% of these were reported from California and lacked information on county that year, making annual rate ratios difficult to compute. These cases are most likely attributable to an epidemic that occurred in Los Angeles [14]. Although reported measles cases declined during 1992 to 540 in the border region, cases there represented a greater than 4- and 10-fold excess, respectively, when compared with measles incidence in regions 2 and 3. By 1998, measles incidence in the United States declined to record lows, yet the 14 cases reported in the border region that year represented a 9- and 5-fold excess, compared with incidences in regions 2 and 3, respectively. The combined incidence of measles, mumps, and rubella throughout the study period was ∼3-fold greater in the border JID 2000;182 (November) US-Mexico Border Infectious Disease Morbidity Figure 3. 1507 Rate ratios of disease incidence in regions 1 and 3, 1990–1998 region than in region 3, even though we excluded all cases in border states with missing county information and, thus, underestimated the true incidence in the border region. Taken together over the study period, the 6 major childhood vaccinepreventable diseases (measles, mumps, rubella, diphtheria, pertussis, and tetanus) occurred in the border region at an incidence rate 85% greater than the incidence in region 3. The age distribution of cases was similar among the 3 regions: measles and pertussis most commonly affected preschool children, mumps most commonly affected older children and adolescents, and rubella affected both young children and adults. Discussion When compared with US nonborder regions, we found significant excess morbidity due to certain infectious diseases in the US region bordering Mexico. This excess morbidity was noted in particular for botulism, brucellosis, diphtheria, hepatitis A, measles, mumps, rabies, rubella, salmonellosis, and shigellosis. Exceptions to this pattern included invasive H. influenzae disease, malaria, meningococcal disease, and trichinosis, which occurred at lower rates in the border region than in the 2 nonborder regions. In contrast to national rates outside the border states, we observed a 13-fold excess in the incidence of hepatitis A in the border region. Hepatitis A occurs worldwide and is often inversely correlated with socioeconomic status, occurring more commonly in areas with poor sanitation conditions [15]. Al- though the United States is generally considered an area of low endemicity for hepatitis A, areas of intermediate and high endemicity exist, as evidenced by recurrent outbreaks of the disease on American Indian reservations [16]. In the United States, rates of hepatitis A are ∼2-fold higher among Hispanics than among non-Hispanics [17]. Several recent studies of hepatitis A in the border region have helped to further clarify the risk factors for infection. In a serosurvey of antibody to hepatitis A virus among adult women in El Paso, Texas, 75.8% of women had evidence of past infection [18]. Past infection was associated with increasing age and with Hispanic ethnicity. A similar serosurvey in a Texas border community found that 16.9% of primary school children were antibody positive for hepatitis A virus [19]. Past infection was positively associated with characteristics of low socioeconomic status (e.g., low maternal education, household crowding, and inadequate water and sewage facilities). Underscoring these results, another study among children in south Texas reported that the independent risk of hepatitis A infection was associated with residence in a colonia [20]. To help reduce the burden of this disease in the border region, routine use of the hepatitis A vaccine is recommended among children in select communities [21, 22]. Our data showed that shigellosis occurred in the border region at an incidence more than double that in region 3. There was a consistent excess incidence of shigellosis in the border region each year of the 1996–1998 period for which complete data were available. Shigellosis is endemic in many parts of 1508 Doyle and Bryan Mexico, and molecular studies of bacterial isolates among shigellosis patients in southern Arizona have suggested an association with travel to Mexico [23]. In addition, travel history to Mexico is associated with resistance to trimethoprim-sulfamethoxazole among Shigella isolates in the United States [24]. Independent of the risks posed by travel to endemic regions in Mexico, the risk of infection with enteric pathogens, such as Shigella species, is undoubtedly increased by the inadequate water and sanitation infrastructure in many colonias and other border communities of the desert Southwest [6, 7]. Future research might appropriately focus on assessing the importance of microbial resistance to antibiotics in the border region. Although rare in the United States, brucellosis occurred in the border region at a rate 8 times the national rate. The epidemiology of human brucellosis in the United States has changed from an occupational disease associated with animal contact to a foodborne disease associated with consumption of unpasteurized dairy products that predominantly affects Hispanics [25–28]. Although our database did not include any information on bacterial species or behavioral risk factors associated with cases, 180% of brucellosis cases in the 4 border states and 58% of cases occurring nationally were among Hispanics. Another rare zoonotic disease, human rabies, also occurred at higher rates in the border region. In the Mexican states bordering the United States, urban epizootics of canine rabies continue to occur [29]. In such settings, children are often at highest risk for exposure to rabies via dog bites. Viral studies of human cases reported from US border states have implicated urban canine rabies strains found primarily in the border region [30]. These same urban canine strains have also been linked to the recent coyote rabies epizootic in southern Texas that resulted in a massive effort to vaccinate coyotes with an experimental oral baiting program [31, 32]. Limited resources and minimal public health infrastructure in some border communities may have hampered local efforts to maintain animal control and canine (dog) rabies vaccination programs. We found that major childhood vaccine-preventable diseases occurred in the border region at nearly twice the incidence in nonborder states. This excess incidence in the border region was driven largely by a 1 4-fold excess in measles incidence, which accounted for most cases in the region for the 6 diseases prevented by the MMR and DPT vaccine series. Nonetheless, this observation suggests that the excess morbidity caused by these readily preventable diseases may have resulted from an inadequate system to ensure routine immunizations in the border region. Between 1986 and 1994, Mexico was the leading source of measles cases imported to the United States, accounting for 31% of all imported cases [33]. Five of the top 20 US counties reporting imported measles were in the border region; 8 of the top 20 were in border states. Imported measles cases may trigger JID 2000;182 (November) a local epidemic in a poorly vaccinated community because of the high infectivity of the virus. During the national measles resurgence witnessed from 1988 to 1991, cases occurred most commonly in unvaccinated preschool children [34, 35]. The national epidemic was characterized by large outbreaks in urban areas, such as San Diego, Los Angeles, Dallas, Houston, Chicago, Milwaukee, and New York. In 1990, 61% of measles cases reported nationally occurred in the states of California and Texas [36], and the border region experienced outbreaks among preschool- and schoolaged children [37]. The large outbreak in Los Angeles continued in 1991, encompassing Riverside County, California, in the border region [14]. Although cases reported nationally declined dramatically during 1992, Hidalgo County in the Texas border region reported a large outbreak that year [38]. Despite improvements in vaccination coverage resulting from the measles resurgence, children living below the poverty line continued to have lower measles, DTP, and oral polio vaccine (OPV) coverage, when assessed immediately after the epidemic period [38]. Three border counties (San Diego; Maricopa, AZ; and El Paso, TX) included in the National Immunization Survey of 1994–1995 demonstrated coverage of 75%–81% for the 4:3:1 dose DTP:OPV:MMR vaccination series [39]. Although measles incidence is currently at historic lows in the United States [40], our analysis suggests that the border region continues to be overrepresented. Prevention of future resurgence of vaccine-preventable diseases in the border region will require ongoing surveillance of vaccination coverage in the region plus innovative outreach methods in communities with inadequate coverage [41]. Efforts to reduce the incidence of vaccine-preventable disease in the region may also benefit from the recent change in Mexican vaccination policy that added MMR to the routine schedule in place of a single-antigen measles vaccine [42]. The NNDSS database used for this study is derived from a passive surveillance system of disease reporting among states. The NNDSS, therefore, likely underestimates the true burden of disease, and reported incidence rates should be interpreted with caution. Notifiable diseases are detected and reported with varying levels of completeness across individual states because of state-specific differences in surveillance methodology. Direct state-to-state comparisons of disease incidence, therefore, are potentially misleading. Rate ratios that compare multistate regions of the country, such as those in this study, may be a more accurate measure of relative morbidity, since the underestimation of disease inherent in NNDSS is less likely to differ in magnitude between regions involving multiple states. Nevertheless, further operational research is needed to evaluate the sensitivity of routine infectious disease surveillance in the border region. In addition, disease rates over many years are influenced by both endemic and epidemic patterns. Regional differences in focal endemicity for diseases such as plague and JID 2000;182 (November) US-Mexico Border Infectious Disease Morbidity large multistate outbreaks of diseases such as salmonellosis must be considered when interpreting these results. This study was limited by missing county information for some diseases during some years. Thus, rate ratios that compare disease incidence in border counties to nonborder regions are potentially biased for the diseases with a high proportion of missing county information. Attempts were made to minimize this bias by modifying the analysis period for diseases limited by missing county information. Our study was further limited by missing data from case reports on age, race, and ethnicity of patients [43]. Therefore, morbidity was generally expressed as crude incidence. Regional differences persisted, however, after stratifying by ethnicity and age group. In conclusion, we found significant excess morbidity due to infectious disease in the US region bordering Mexico, compared with nonborder regions. In particular, this excess morbidity was noted for diseases associated with poor sanitation, zoonotic diseases, and vaccine-preventable diseases. Inadequate public health infrastructure is likely an important underlying cause for the excess morbidity observed. An improved water and sanitation infrastructure is needed to address the excess burden of enteric disease. Enhanced surveillance is recommended to more accurately monitor enteric and zoonotic disease incidence and vaccination coverage levels. Finally, a binational approach to surveillance and prevention is needed to address public health problems that defy international boundaries. Acknowledgments We thank Carol Knowles, Willie Anderson, and Man-huei Chang for assistance in data management; Pani Ellinas and Douglas Thoroughman for helpful comments and suggestions; and Stephen Waterman, Samuel Groseclose, and Kate Glynn for critical review of the manuscript. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. References 1. Warner DC. Health issues at the US-Mexican Border. JAMA 1991; 265: 242–7. 2. 103rd Congress. Public law 103-400 [S.1225]: United States–Mexico Border Health Commission Act, Washington, DC: US Government Printing Office, October 1994. 3. US Census Bureau, Housing and Household Economics Division. CPH-L184. The 100 poorest counties in the United States. Washington, DC: US Government Printing Office, 1989. 4. Jones DB. Trouble on the border: international health problems merge at the Rio Grande. Tex Med 1989; 85:28–33. 5. Davidhizar R, Bechtel GA. Health and quality of life within colonias settlements along the United States and Mexico border. Public Health Nurs 1999; 16:301–6. 6. Mroz RC, Morales LL, VanDerslice J. Health and hygiene in the colonias: water and disease. Fam Commun Health 1996; 19:49–58. 7. Mroz RC, Pillai SD. Bacterial populations in the groundwater on the USMexico border in El Paso County, Texas. South Med J 1994; 87:1214–7. 8. Cech I, Essman A. Water sanitation practices on the Texas-Mexico border: implications for physicians on both sides. South Med J 1992; 85:1053–64. 9. Roush S, Birkhead G, Koo D, Cobb A, Fleming D. Mandatory reporting 24. 25. 26. 27. 28. 29. 30. 31. 32. 1509 of diseases and conditions by health care professionals and laboratories. JAMA 1999; 282:164–70. Cates W, Williamson G. Descriptive epidemiology: analyzing and interpreting surveillance data. In: Teutsch S, Churchill R, eds. Principles and practices of public health surveillance. New York: Oxford University Press, 1994: 96–135. US Census Bureau, Population Estimates Program. CO-98-11 and ST-98-26. Population estimates by race and Hispanic origin: 1 July 1994. Washington, DC: US Government Printing Office, 1999. US Census Bureau, Population Estimates Program. CO-98-13. Population estimates by age group: 1 July 1994. Washington, DC: US Government Printing Office, 1999. Angulo FJ, Getz J, Taylor JP, et al. A large outbreak of botulism: the hazardous baked potato. J Infect Dis 1998; 178:172–7. Centers for Disease Control and Prevention. Measles surveillance: United States, 1991. MMWR Morb Mortal Wkly Rep CDC Surveill Summ 1992; 41(SS-6):1–12. Shapiro CN, Margolis HS. Worldwide epidemiology of hepatitis A virus infection. J Hepatol 1993; 18:S11–4. Shaw FE, Shapiro CN, Welty TK, Dill W, Reddington J, Hadler SC. Hepatitis transmission among the Sioux Indians of South Dakota. Am J Public Health 1990; 80:1091–4. Shapiro CN, Coleman PJ, McQuillan GM, Alter MJ, Margolis HS. Epidemiology of hepatitis A: seroepidemiology and risk groups in the USA. Vaccine 1992; 10:S59–62. Redlinger T, Nickey L, O’Rourke K, Martinez G. Elevated hepatitis A and E seroprevalence rates in a Texas/Mexico border community. Tex Med 1998; 94:68–71. Redlinger T, O’Rourke K, VanDerslice J. Hepatitis A among schoolchildren in a US-Mexico border community. Am J Public Health 1997; 87:1715–7. Leach CT, Koo FC, Hilsenbeck SG, Jenson HB. The epidemiology of viral hepatitis in children in south Texas: increased prevalence of hepatitis A along the Texas-Mexico border. J Infect Dis 1999; 180:509–13. Centers for Disease Control and Prevention. Prevention of hepatitis A through active or passive immunization: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Morb Mortal Wkly Rep 1999; 48(RR-12):1–37. Bell BP, Shapiro CN, Alter MJ, et al. The diverse patterns of hepatitis A epidemiology in the United States: implications for vaccination strategies. J Infect Dis 1998; 178:1579–84. Litwin CM, Ryan KJ, Chipowsky S, Storm A, McCombie S. Molecular epidemiology of Shigella sonnei in Pima County, Arizona: evidence for a Mexico-related plasmid. J Infect Dis 1990; 161:797–800. Tauxe RV, Puhr ND, Wells JG, Hargrett-Bean N, Blake PA. Antimicrobial resistance of Shigella isolates in the USA: the importance of international travel. J Infect Dis 1990; 162:1107–11. Eckman MR. Brucellosis linked to Mexican cheese. JAMA 1975; 232:636–7. Young EJ. Brucellosis outbreak attributed to ingestion of unpasteurized goat cheese. Arch Intern Med 1975; 135:240–3. Taylor JP, Perdue JN. The changing epidemiology of human brucellosis in Texas, 1977–1986. Am J Epidemiol 1989; 130:160–5. Chomel B, DeBess E, Mangiamele D, et al. Changing trends in the epidemiology of human brucellosis in California from 1973 to 1992: a shift toward foodborne transmission. J Infect Dis 1994; 170:1216–23. Eng T, Fishbein D, Talamante H, et al. Urban epizootic of rabies in Mexico: epidemiology and impact of animal bite injuries. Bull World Health Organ 1993; 71:615–24. Noah DL, Drenzek CL, Smith JS, et al. Epidemiology of human rabies in the United States, 1980 to 1996. Ann Intern Med 1998; 128:922–30. Clarke K, Neill S, Smith J, Wilson P, Whadford V, McKirahan G. Epizootic canine rabies transmitted by coyotes in south Texas. J Am Vet Med Assoc 1994; 204:536–40. Fearneyhough M, Wilson P, Clark K, et al. Results of an oral rabies vaccination program for coyotes. J Am Vet Med Assoc 1998; 212:498–502. 1510 Doyle and Bryan 33. Vitek CR, Redd SC, Redd SB, Hadler SC. Trends in importation of measles to the United States, 1986–1994. JAMA 1997; 277:1952–6. 34. Dales L, Kizer K, Rutherford G, Pertowski C, Waterman S, Woodford G. Measles epidemic from failure to immunize. West J Med 1993; 159:455–64. 35. Atkinson WL, Orenstein WA, Krugman, S. The resurgence of measles in the United States, 1989–1990. Annu Rev Med 1992; 43:451–63. 36. Centers for Disease Control and Prevention. Measles: United States, 1990. MMWR Morb Mortal Wkly Rep 1991; 40:369–72. 37. Hutchins S, Markowitz L, Atkinson W, Swint E, Hadler S. Measles outbreaks in the United States, 1987 through 1990. Pediatr Infect Dis J 1996; 15: 31–8. 38. Centers for Disease Control and Prevention. Measles: United States, 1992. MMWR Morb Mortal Wkly Rep 1993; 42:378–81. 39. Centers for Disease Control and Prevention. National, state, and urban area JID 2000;182 (November) vaccination coverage levels among children aged 19–35 months: United States, April 1994–March 1995. MMWR Morb Mortal Wkly Rep 1996; 45:145–50. 40. Centers for Disease Control and Prevention. Epidemiology of measles: United States, 1998. MMWR Morb Mortal Wkly Rep 1999; 48:749–53. 41. Waterman S, Hill L, Robyn B, et al. A model immunization demonstration for preschoolers in an inner-city barrio, San Diego, California, 1992–1994. Am J Prev Med 1996; 12:8–13. 42. Santos Preciado JI. Nuevo esquema de vacunacion en Mexico. Salud Publica Mex 1999; 41:1–2. 43. Centers for Disease Control and Prevention. Reporting race and ethnicity data: National Electronic Telecommunications System for Surveillance, 1994–1997. MMWR Morb Mortal Wkly Rep 1999; 48:305–12.