ePortfolio: Association between Tuberculosis and Diabetes in Texas

Association Between Tuberculosis And Diabetes In The Mexican Border And Non-Border Regions of Texas Full Text


A study of the association between Tuberculosis and Diabetes, in the Mexican border and non-border regions of Texas; Performed by The University of Texas, and published by the American Society of Tropical Medicine and Hygiene. History shows a positive association between incident rates of Diabetes and Tuberculosis, specifically in populations with low socio-economic status and high immigration rates.

Three scientists at the University of Texas; Adriana Perez, H. Shelton Brown III, and Blanca I. Restrepo, sought to find if there was an association between Tuberculosis and Diabetes. They set their study in the state of Texas with specific interests on the regions bordering Mexico. They began with these 2 facts that propagated their study: History has shown a positive association between Diabetes and Tuberculosis, as areas with low socio-economic status have high incidence rates for both diseases; 14 of the border regions on the Texas-Mexico border are among the poorest in the United States. In 2002 the Tuberculosis incidence rate in the United States was at an all time low with only 5.6 occurrences per 100,000 people, whereas in Mexico this rate was nearly double at 10.1 per 100,000 people. These incidence rates dramatically spike as we decrease our scope and zoom in on our zone of interest, within the state of Texas, 7.2 cases per 100,000 people, and the border regions are at 13.1 cases per 100,000 people. Two cities on the border of Texas that have the highest incidence rates are McAllen and Brownsville, whose rates roll in at 12.8 and 17.4 per 100,000 people, their sister cities on the Mexican side of the border Reynosa and Matamoros, spike at 43.9 and 70.3 per 100,000 people.
Both Diabetes and Tuberculosis are increasingly prevalent in Texas, and both are ongoing major public health problems. The large rates of Mexican immigrants that live in Texas have posed a large problem with control of the spread of Tuberculosis – in 2002, 51.6% of the cases in Texas were born in the United States, 24.6% were from Mexico, and 17.8% were from other countries. As more people immigrate between these areas the higher risk of exposure to the bacterium that causes Tuberculosis, may continue the upward association and trend between Diabetes and Tuberculosis.

Materials + Methods

A case-control analysis was done with cross-sectional data from Texas’ database of hospital discharges from 1999-2001. For this study they separated the state into 2 groups, the 15 border counties {Brewster, Cameron, El Paso, Hidalgo, Hudspeth, Jeff Davis, Kinney, Maverick, Presidio, Starr, Terrell, Val Verde, Webb, Willacy, and Zapata} and the remaining 239 “non-border” counties. To determine the region these patients were from, patient’s data was classified by zip code, county code and state code. Every individual that was discharged from a facility with a diagnosis of Tuberculosis {via ICD-9 codes} were admitted into the study. Many co-variables were extracted via individual’s demographics {age, sex, race, and insurance type} and any comorbidity factors {diabetes, renal failure, alcohol/drug use, cancer, surgeries, and nutritional deficiencies}. All patients were 15 years are older, and every readmission was listed as a new patient variable within the study. Income and education were taken from the US census from 2000 for each zip code area, and adjusting for socio-economic status via the 1999 median household income by zip codes.
Four thousand nine hundred and fifteen Tuberculosis hospitalizations were tracked from 1999-2001; with 1,244 in border regions and 3,671 non-border regions of Texas. The three main reasons for hospitalization of these cases contained: Tuberculosis infection, Pneumonia, and rehabilitation care. The control group {patients without diagnoses’ of Tuberculosis} contained 70,808 with 12,563 from the border regions and 58,245 individuals from the non-border regions. The three main reasons for hospitalization of these cases contained: acute appendicitis, acute pulmonary heart disease, and phlebitis/thrombophlebitis.


– Hospitalized Tuberculosis cases more likely to come from neighborhoods with lower median incomes in all regions of Texas.

– Tuberculosis cases were less likely to be located in areas with high percentage of higher education graduates for both border and non-border regions.

– Patients from the border were more likely to have co-morbidities of diabetes, chronic renal failure, or have a nutritional deficit, and less likely to have cancer than the controls.

– Hospitalized Tuberculosis cases are more likely to be Hispanic men ≥ 45 years.

– Tuberculosis patients were 10x as likely to be alcohol users compared with the controls.

– Having Medicare/Medicaid or private insurance was associated with a lower risk in both border and non-border regions.

– Federal insurance/ VA/ or military insurance was associated with a higher risk in the non-border regions.

– Border region patients with Diabetes had more that twice the risk of Tuberculosis than those without diabetes.

– Non-border patients with Diabetes had 1.5 times the risk of Tuberculosis as those without Diabetes.

– Border patients with Diabetes had a 1.3 times the risk of Tuberculosis that those who lived in non-border regions.

– Hispanics with diabetes have over 2 times the risk of Tuberculosis as those without diabetes.

– Those with diabetes and without insurance have a greater risk of developing Tuberculosis than those who only have exposure.

– Nutritional deficits lead to decreased immune ability and increased risk of Tuberculosis.


Diabetes is a risk factor for developing Tuberculosis, and those living close to the Texas border are at a high risk for exposure due to the rate of those living in the area with latent Tuberculosis infections. The Hispanic population has an opposite effect as those living in the non-border regions are 23% more likely to develop Tuberculosis than those living in the border regions, as most Hispanic are immigrants from areas that are Tuberculosis endemic countries and are likely to have latent Tuberculosis infections. Those with latent Tuberculosis infections have a higher probability of reactivation of this infection with the development of Diabetes. These numbers will continue to grow as we see an ever-looming number of Diabetes rates increasing year after year.