Geospatial analysis as a tool to identify target areas for Chagas disease education for healthcare providers

Abstract

Chagas Disease (CD) is a neglected zoonotic disease of the Americas. It can be fatal if not diagnosed and treated in its early stages. Using geospatial and sensitivity analysis, this study focuses on understanding how to better allocate resources and educational information to areas in the United States, specifically Texas, that have the potential for increased risk of CD cases and the associated costs of addressing the disease. ICD-9 and 10 inpatient hospital diagnostic codes were used to illustrate the salience of potentially missed CD diagnoses (e.g., cardiomyopathic diagnoses) and where these are occurring with more frequency. Coding software along with GIS and Microsoft Excel 3D mapping were used to generate maps to illustrate where there may be a need for increased statewide surveillance and screening of populations at greater risk for CD. The CD cases reported to the Texas Department of State Healthcare Services (TxDSHS) are not homogenously dispersed throughout the state but rather, reveal that the incidences are in clusters and primarily in urban areas, where there is increased access to physician care, CD research and diagnostic capabilities.

Full text

Pachecoetal. BMC Infectious Diseases (2022) 22:590
https://doi.org/10.1186/s12879-022-07577-y
RESEARCH
Geospatial analysis asatool toidentify
target areas forChagas disease education
forhealthcare providers
Gerardo J. Pacheco1*, Lawrence Fulton1, Jose Betancourt1, Ram Shanmugam1 and Paula Stigler Granados2
Abstract
Chagas Disease (CD) is a neglected zoonotic disease of the Americas. It can be fatal if not diagnosed and treated in
its early stages. Using geospatial and sensitivity analysis, this study focuses on understanding how to better allocate
resources and educational information to areas in the United States, specifically Texas, that have the potential for
increased risk of CD cases and the associated costs of addressing the disease. ICD-9 and 10 inpatient hospital diag-
nostic codes were used to illustrate the salience of potentially missed CD diagnoses (e.g., cardiomyopathic diagnoses)
and where these are occurring with more frequency. Coding software along with GIS and Microsoft Excel 3D map-
ping were used to generate maps to illustrate where there may be a need for increased statewide surveillance and
screening of populations at greater risk for CD. The CD cases reported to the Texas Department of State Healthcare
Services (TxDSHS) are not homogenously dispersed throughout the state but rather, reveal that the incidences are in
clusters and primarily in urban areas, where there is increased access to physician care, CD research and diagnostic
capabilities.
Keywords: Chagas disease, Trypanosoma cruzi, American Trypanosomiasis, Neglected tropical diseases, Geospatial
analysis, Heart disease
© The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which
permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the
original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or
other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this
licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecom-
mons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Introduction
Chagas Disease (CD) is a neglected zoonotic disease [1]
of the Americas that can be fatal if left undiagnosed and
not treated in its early stages. CD accounts for the high-
est burden of any parasitic disease in the Latin Ameri-
can countries where it is endemic. Trypanosoma cruzi,
the parasite responsible for Chagas disease, is endemic
throughout Central and South America and is also found
in North America, including Mexico and the Southern
United States (U.S.) [2–5]. An estimated 8 million peo-
ple in Latin America have CD [6]. Over 28,000 people are
infected each year in Mexico, Central America and South
America, accounting for at least 12,000 deaths per year
[7]. It is estimated that there are approximately 326,000–
347,000 Latin American-born infected individuals living
in the U.S., however the number of autochthonous cases
is unknown [8].
Reduviids, also known as triatomines or "kissing bugs",
are blood-feeding insects that are the primary vector for
CD transmission. Transmission generally occurs when
an infected triatomine defecates near the bite site and
the feces enters the wound, transmitting the parasite
into the host. Triatomines may transmit the parasite to
mammals, including humans [1], but can also infect res-
ervoir hosts such as canines, opossums, raccoons, and
other domestic [9, 10], and sylvatic animals [11]. Other
transmission routes include oral ingestion of triatomine
feces contaminated food or drink, congenital transmis-
sion from mother to a fetus, exposure to contaminated
Open Access
*Correspondence: gjp46@txstate.edu
1 School of Health Administration, Texas State University, San Marcos, TX
78666, USA
Full list of author information is available at the end of the article
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Page 2 of 11
Pachecoetal. BMC Infectious Diseases (2022) 22:590
blood products and through organ transplantation from
an infected donor [4]. Blood donation screening is the
most common means by which individuals learn about
their CD diagnosis in the U.S. [12].
CD includes two main phases: acute and chronic [1,
13]. Acute infections occur up to the first two months
of the initial infection, which may manifest with mild
flu-like symptoms or prolonged febrile illness [14, 15].
Other symptoms may include malaise; enlarged spleen,
liver, and lymph nodes; localized or generalized edema;
and chagomas or breaks in the skin [1]. Infection may
also result in abnormal electrocardiogram (ECG) read-
ings [10]. Acute infection may manifest as early as one-
week post- exposure and may be self-limiting in most
individuals [10]. e patient may not seek medical atten-
tion since the symptoms are mild and not unique to CD.
During the chronic stage, two presentations are possible:
the indeterminate form and determinate form. e inde-
terminate form is characterized as a chronic infection by
T. cruzi without specific organ damage and is commonly
asymptomatic. e determinant form is characterized as
having specific organ damage with complications, which
may include cardiac manifestation (e.g., cardiomyopathy,
heart failure, altered heart rate or rhythm) and intestinal
complications [13]. Approximately 70–80% of infected
individuals [13, 16] will transition from the acute phase
and remain in a latent or indeterminate chronic form of
the disease (mostly asymptomatic), which may persist
as a lifelong infection [4]. e danger of this asympto-
matic status is that once symptoms do manifest, elimi-
nating the parasite is either more difficult or impossible
with the latter case resulting in death. Treatment of the
disease is with anti-parasitic drugs (Benznidazole or
Nifurtimox) [17]; however research does not conclusively
suggest a reduction in mortality after treatment nor a
reversal of symptoms if the patient has entered into the
chronic phase with specific organ damage [18]. Approxi-
mately 20–30% of infected individuals will progress from
the indeterminate chronic phase without organ damage
to a “clinically evident disease” or chronic determinate
phase. is progression occurs months to decades after
becoming infected [15]. Infection in humans can present
in many forms such as non-ischemic cardiomyopathy,
heart failure, cardiac arrhythmias and sometime gastro-
intestinal disease [10]. Cardiovascular pathophysiology is
believed to be multi-causal (e.g., direct parasitic aggres-
sion [19–22]). Sudden death due to cardiac complications
can also occur [6]. For the scope of this study, heart-
related symptoms were the primary focus.
Chagasic cardiomyopathy may include cardiac arrhyth-
mias, heart failure, and risk of sudden death from ven-
tricular fibrillation or tachycardia or thromboembolic
events [10, 23]. Cardiovascular disease in CD patients
is believed to be the result of the presence of the para-
site in the cardiac tissue causing an immune-mediated
myocardial injury [24]. A recent study by Hyson et al.,
screened 1156 patients for CD and revealed that out of
the 23 patients that had positive serological screenings,
cardiomyopathy and congestive heart failure was present
in 43% of the cases [25]. CD may present as an idiopathic
cardiomyopathy and therefore be overlooked by many
or most healthcare providers as a diagnosis if they are
inexperienced in seeing patients with CD. An estimated
10–15% of the total U.S. population (or 30,000 to 45,000
individuals) is estimated to be living with undiagnosed
CD cardiomyopathy [26]. Many U.S. physicians and other
healthcare providers (HCPs) are not well versed in CD
screening, diagnostics, or treatment [27].
Between 2013 and 2019, 184 total cases of CD were
reported in Texas [28]. Although regarded typically as a
rare neglected tropical disease [1], current vector surveil-
lance, [29, 30] the increased frequency of Chagas positive
blood donors, [31–33] and population migration, [34,
35] demonstrate why more CD cases may be going unde-
tected in the U.S.
With regards to U.S. physicians’ knowledge of CD,
recent pre-post evaluation of healthcare providers
(HCPs) in Texas suggests specific gaps in medical train-
ing and awareness on screening and diagnosing patients
[36]. A separate study using a mixed methods approach
supports these findings: HCPs were not confident overall
in their skills to screen, diagnose, and treat CD patients
[37]. In a recent study of sampled U.S. obstetricians, less
than one-third of the study sample knew of the testing
protocol, and one fifth knew of the follow-up protocols
once a patient received a positive diagnosis [38]. us,
facilitating CD education to healthcare providers remains
a challenge [27, 36].
is study aims to illustrate potential heart-related
CD in Texas using a geospatial and sensitivity analysis
to inform health policy makers of the areas where pre-
vention, education and screening efforts might be best
served. Given the risk factors for exposure and transmis-
sion to CD and etiology of specific strains (i.e., heart-
related symptoms), we hypothesize an increased number
of suspected cases of CD throughout the state among
younger Hispanic individuals. Based on the suspected
case definitions (described below), we expect a higher
proportion of individuals with undiagnosed CD with
heart-related symptoms.
Methods
Data sources
e Inpatient Public Use Data File (IPUDF) for 2016,
maintained by the Texas Department of State Health
Services (TxDSHS), and the number of Chagas cases
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Page 3 of 11
Pachecoetal. BMC Infectious Diseases (2022) 22:590
confirmed for 2019 by TxDSHS were used for geospa-
tial analysis [39, 40]. In Texas, CD became a reportable
condition in 2013; therefore hospital inpatient data from
2013 to 2016 were acquired. In Texas, a CDC confirmed
diagnosis is reported to TxDSHS and is listed as either
acute, chronic indeterminant or chronic determinant
based on epidemiological investigations and interviews
conducted by the local or regional health departments.
Census data [41] was used to download the American
Community Survey (ACS) 5-year Texas population esti-
mates for 2016. is included Texas demographic data
on age and Hispanic status by county. A base map was
created by downloading the shapefile for the Texas coun-
ties from the U.S. Census Bureau (i.e., TIGER/Line Web
interface) [42].
Variables andcase denitions
Patient demographics and patient diagnostic codes origi-
nated from the inpatient data (i.e., the raw quarterly base
files from the PUDF, 2013 to 2016). e demographic
variables included: patient’s age group (i.e., < 18; 18–44;
45–64; 65–74; and 75); ethnicity (Hispanic or non-His-
panic); race (American Indian/ Eskimo; Asian or Pacific
Islander; Black; White; or Other); and sex code (male or
female). Additional variables that were kept from the ini-
tial raw inpatient PUDF dataset included: record identi-
fication number for each hospital admission; patient’s
county and zip code of residence; provider ID; and type
of admission.
e admitting diagnosis and the twenty-four principle
diagnostic codes were the queried variables from each
quarterly base data file (e.g., IPUDF) for each hospital
admission. e process was iterative and exploratory in
identifying both the diagnosed CD cases and the poten-
tially missed CD cases (undiagnosed) manifested through
heart complications. ese diagnoses variables were
re-coded to determine if that particular patient record
contained the ICD 9 or ICD 10 diagnostic codes of inter-
est (e.g., Table1: CD cases). A total of 3,088,978 hospi-
tal inpatient records were identified once all the datasets
were combined (i.e., 2013–2016). A dichotomous varia-
ble (i.e., for any admission that had any of the CD-related
ICD diagnosis codes) was created using the CD cases.
A cardiologist with expertise in diagnosing CD was
consulted with to identify and further review heart-
related diagnostic codes (i.e., the potentially-missed CD
cases). e list of ICD codes was then reviewed by a sec-
ond cardiologist with experience in diagnosing and treat-
ing CD patients to eliminate unnecessary codes. e ICD
diagnostic code and corresponding definition for heart-
related cases, a proxy for potentially-missed CD diagno-
sis, are shown on the second half of Table1. Additional
ICD-10-CM codes for heart-related diagnoses were
identified to expand the list of potentially-missed cases
(Table2). e IPUDF transitioned from utilizing ICD-9
diagnosis codes to ICD-10 in 2016. Additionally, this was
the most recent year of available data for the analyses at
Table 1 Chagas disease and potential cardiomyopathy-related ICD 9 and ICD 10 codes with descriptions
Case ICD Version Diagnostic code Description
Chagas disease ICD-9-CM 086.0 Chagas with heart involvement
ICD-9-CM 086.1 Chagas with other organ involvement
ICD-9-CM 086.2 Chagas without mention of organ involvement
ICD-10-CM B57.0 Acute, heart
ICD-10-CM B57.1 Acute, without heart
ICD-10-CM B57.2 Chronic, with heart
Potentially missed CD cases (heart-
related codes) ICD-9-CM 414.8 Other forms of chronic ischemic heart disease
ICD-9-CM 422.91 Idiopathic myocarditis
ICD-9-CM 425.8 Cardiomyopathy, excludes Chagas
ICD-9-CM 425.4 Cardiomyopathy, includes idiopathic
ICD-10-CM I25.5 Ischemic cardiomyopathy
ICD-10-CM I42.9 Cardiomyopathy, unspecified
Table 2 Additional ICD-10 cardiomyopathy codes that could be
potentially related to a missed Chagas disease diagnosis
Diagnostic code Description
I428 Other cardiomyopathies
I429 Cardiomyopathy, unspecified
I4510 Unspecified Right Bundle Branch Block
I452 Bifascicular block
I441 Atrioventricular block, second degree
I442 Atrioventricular block, complete
I472 Ventricular tachycardia
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Page 4 of 11
Pachecoetal. BMC Infectious Diseases (2022) 22:590
the time of the study. e expanded list was only used to
query 2016 to identify the potentially-missed CD cases.
A dichotomous variable (i.e., for any admission that had
any of the heart-related ICD diagnosis codes) was created
using the heart-related ICD codes.
Data collection andmanagement
e 2016 quarterly IPDUF base files were individu-
ally exported as a comma separated value file (CSV) to
Excel. Exploratory data analysis and data cleaning were
performed, and patient records that did not contain the
case definitions were eliminated from the dataset. e
combined raw IPUDF contained over 3 million hospital
admissions of which 3.1% were included in this study,
as they contained a heart related/CD diagnosis code.
Demographic data for the state of Texas were down-
loaded using the American Fact Finder web application.
e ACS estimates were chosen for the Hispanic popu-
lation and age categories. Microsoft Excel was used for
data cleaning. To calculate the Hispanic proportion, the
number of Hispanics was divided by the total population
for each county. To create the table for age groups, only
the population estimates for males, females, and all, aged
20–59 were calculated.
Mapping
Once definitions were finalized, the dichotomous
variables for CD cases and potentially- missed CD
(heart-related) were compiled, aggregated by year, and
tabulated. ArcMap GIS (Version 10.6.0) was used to visu-
alize the variables [43]. is software coupled with Excel
2016’s innate 3d mapping provided geographic descrip-
tive capability. A series of maps were created to explore
geographic areas of interest.
Modeling
All but two of the 254 counties in the state were included
in the models. Two of the 254 Texas counties were
excluded in the models due to low reporting and patient
county suppression, “e county code is suppressed if
a county has fewer than five discharges for that quar-
ter” [40]. Linear regression with residuals was applied
using the following variables: male population, Hispan-
ics, race (Asian or Pacific Islander; Black, White, and
Other), county population, population density (persons
per km2, and rate). e modelling techniques (e.g.,
Queen’s, Global Moran’s index, etc.) used can be refer-
enced online: https:// rpubs. com/R- Minat or/ chaga s1.
For instance, Moran’s index I captures (Li etal. 2007) the
intercorrelation between two adjacent geographic units
(counties or towns etc.) with respect to the CD incidence.
e expected value of the Moran’s index is −1/(N−1)
under the assumption of no geo spatial autocorrelation
among the units. Where N is the number geographi-
cal units. An analysis on Global Moran’s I was used to
account for location and variable differences [44]. Robust
linear mixed-method models with Lagrange multiplier
diagnostics as a way to test for spatial dependence [45]. A
generalized spatial two-stage least squares model (using
STSLS) was used. is model with weighted matrix
accounted for variability. e model controlled for gen-
der (males), age, and population.
Results
Descriptive statistics Chagas diagnoses andheart‑related
ICD codes
Utilizing the IPDUF database, 98 CD diagnoses between
2013 and 2016 were identified. Of these 98 cases, 78 pre-
sented with heart involvement and the other 20 were
not specified. ere were 366,575 cases that fit the ICD
9 or 10 definitions for potentially missed diagnosis (i.e.,
heart-related diagnoses). “Cardiomyopathy, including
idiopathic” and “other chronic ischemic heart disease
diagnoses” accounted for the most instances (118,206
and 150,207, respectively). e least occurring diagnos-
tic code was for idiopathic cardiomyopathy (n = 384).
Table3 provides the descriptive statistics for the ICD-
related data: total CD diagnoses and suspected (i.e.,
heart-related conditions).
e data for the suspected cases (i.e., potentially
missed CD cases) were zero-inflated. Of the 254 coun-
ties, only 15 were non-zero. e average number of CD
cases in any county was 0.39, and the median was 0. e
maximum number of confirmed CD inpatient cases was
26 (Dallas County). For total heart-related diagnoses
(i.e., potentially missed CD), only 1 observation had zero
cases (Loving County). e mean was 1443 cases with
a median of 340.5 (large positive skewness). Within our
sample, 49% were male and 39% were Hispanic. e larg-
est Hispanic population resided in Starr County (99%).
Table 3 Descriptive statistics for Chagas disease cases and potentially missed Chagas disease cases using the ICD-related data from
inpatient hospital records in Texas, 2013–2016
Counties Mean SD Median Maximum Total
CD cases 254 0.39 2.46 0 26 98
Potentially Missed CD cases (i.e., Heart-
Related Conditions) 254 1443.21 4998.23 340.50 59,118 366,575
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Page 5 of 11
Pachecoetal. BMC Infectious Diseases (2022) 22:590
e smallest Hispanic population resided in Roberts
County (3%). e median age was 34years old.
Maps
Figure1 illustrates the number of CD cases reported to
TxDSHS between 2013 and 2016 in Texas, by transmis-
sion type as confirmed by the Zoonosis Control Divi-
sion. is map provides a baseline for understanding
geographic concentrations of confirmed cases of CD as
reported by TxDSHS. A total of 91 individual cases were
confirmed and reported, with each corresponding sym-
bol representing a case. Based on the data presented in
this figure, the largest clusters of imported cases were in
Harris and Dallas counties. Individual imported cases
were reported in Potter and Wilbarger counties in the
north, El Paso in the far west, and Shelby and Anderson
counties towards the east. Locally acquired cases were
reported in Bexar County and some in South Texas coun-
ties of Hidalgo, Brooks, and Cameron. It should be noted
that it is possible that locally-acquired cases could be
congenital transmission. However, there is research and
surveillance that indicates positive triatomines in South
Texas have been frequently found near human dwellings,
therefore autochthonous transmission is plausible [4, 46,
47].
Figure 2 is a heat map of the ICD codes for heart-
related diagnosis (i.e., potentially missed CD cases). e
heat map provides color-changing values that indicate
intensity levels of the disease processes (blue being the
least intense and red being the most). From the map, it is
clear that Houston has the most coded cases with Dallas,
Fort Worth, and San Antonio ranking 2nd through 4th,
respectively.
Figure3 represents the ACS estimate of the Hispanic
population. A graduated symbology (i.e., red triangles) is
shown in both maps to illustrate the potential for missed
CD disease diagnosis (displayed as the CD heart related
diagnostic codes) given that a younger population might
be experiencing heart complications. Counties with a
high Hispanic population (75% to 99%) had a range of
heart-related frequency, though the highest numbers
were in El Paso, Maverick, Webb, Hidalgo, and Cameron.
Counties with a lower proportion of a Hispanic popula-
tion such as Randall or Montgomery had large (but not
Fig. 1 Chagas disease cases and transmission categories reported by the Texas Department of State Health Services by county of transmission from
2013 to 2016
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Page 6 of 11
Pachecoetal. BMC Infectious Diseases (2022) 22:590
the highest) numbers of possible CD heart-related diag-
nostic codes. e increase in the triangle symbology in
the population map can be seen in some counties with a
large proportion of the population aged 20 to 59. In six
counties with the largest proportion 55% to 65% in this
age group (i.e., Hartley, Childress, King, Garza, Sterling,
and Concho) had 100 or less heart-related diagnostic
codes.
Figure4 shows the comparison between the CD diag-
nostic codes and the heat map for the number of heart-
related codes. Harris and Dallas/Tarrant, Travis, Bexar,
and Cameron counties show clusters of both ICD-coded
CD and higher proportions of heart-related diagnostic
codes. Wilbarger County near the Texas-Oklahoma bor-
der, a county with 28% Hispanic population, experienced
six CD diagnoses yet only 345 heart-related codes.
Statistical modeling
e mean number of total missed diagnoses was 347.6
with a SD of 1158. e descriptive statistics are presented
in Table4 for the additional variables. e first model
(LM) resulted with an f statistic of 2.93, a coefficient
determination of 0.07755 and a statistical significance
(p-value < 0.05). An interactive map (https:// rpubs. com/
R- Minat or/ chaga s1) was created using the model that
shows the suspected rate (i.e., missed non-ICD Chagas
diagnosis) per 100,000 of the county population.
Discussion
Interestingly, the findings in Fig.4 are congruent with a
previous study of Texas Blood donors, highlighting sig-
nificant localities of concern [33].CD is more prevalent
among individuals who have lived in endemic regions of
Latin America and are considered to be at a higher risk
for CD. However, it is estimated that only about 1% of
the individuals living with CD in the U.S. are aware of
their diagnosis [48]. In our study, heart disease symp-
toms that are common for patients with CD are used as
a proxy to estimate possible cases and indicate areas to
increase screening efforts. Areas in Texas with younger
Fig. 2 Heat map of ICD heart related codes that could be a possible missed Chagas disease diagnosis, 2013–2016
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Page 7 of 11
Pachecoetal. BMC Infectious Diseases (2022) 22:590
Hispanic populations and increased persons presenting
with heart disease that could be related to CD was the
focus of this paper [33]. It should be noted that this is not
the only population at risk in Texas, however they are at
a higher risk than other populations [48]. Presenting data
in a visual and spatial format can often be useful in illus-
trating and contextualizing environmental factors [49].
Given that the CD vector is found throughout the State
of Texas [50], local transmission is not well understood
and the large Hispanic population of Texas, visualization
of potential missed diagnoses is an important and signifi-
cant exploratory analysis. In turn, through GIS analysis
and data visualization, educational and outreach efforts
can be further targeted throughout the state. e applica-
tion of contextualizing maps and integrating with statisti-
cal methods to enhance public health activities has been
described in chronic diseases [51]. However, the value
and use of GIS to inform public health education activi-
ties on CD has not been previously studied.
e CD cases reported to TxDSHS are not homog-
enously dispersed throughout the state but occur in
clusters and primarily in urban areas, where presum-
ably there is increased access to physician care and larger
populations. e policy implication is that screening for
CD should begin with the populations most likely at risk
[48, 52]. e data from TxDSHS [40] show the possibil-
ity for locally acquired or imported infection. Pockets
of locally acquired cases were reported specifically in
Bexar, Hidalgo, Brooks, and Cameron counties. How-
ever, no other areas, (i.e., the panhandle; western Texas
including the El Paso region; and the eastern parts) show
locally acquired infections. Moreover, five newly diag-
nosed CD patients are described in a case report [29]: All
of the patients acquired CD locally and resided in rural
Southeast Texas counties. is highlights the possibil-
ity of persons currently not knowing that they have CD
because many cases remain undiagnosed, particularly
since the disease can become latent. In addition to local
transmission, Texas presents the opportunity to surveil
and diagnose imported cases. It is imperative for HCP’s
throughout the state to recognize CD and be able to
screen and diagnose patients.
Some counties with a high burden of heart-related
diagnosis are also areas with CD diagnosis. e congru-
ence in the urban hubs (Bexar, Dallas, and Harris coun-
ties) reflects the overall population but may also reflect
Fig. 3 Potentially missed Chagas disease cases using heart-related ICD diagnosis codes from 2013 to 2016 and Hispanic population in Texas 2016
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Page 8 of 11
Pachecoetal. BMC Infectious Diseases (2022) 22:590
the availability and ability of physicians in those coun-
ties to recognize CD and appropriately screen, diagnose,
and treat. Conversely the modeling provided new insight
into geographic areas (i.e., Kenedy County that was not
believed by the researchers to be an area if interest).
More focused education and outreach could be targeted
to healthcare providers who may have limited knowledge
in screening and diagnosis of CD in geographic areas
where we find that there could be a higher risk for CD
along with noted elevations in heart disease among a
higher population of Latinos. As of 2017, the estimated
seroprevalence of CD in Mexico was 2.26%, much higher
than previously thought [29]. A study in Starr County,
Texas which lies adjacent to the Mexican border found
that eight of 1196 study participants (0.7%) screened pos-
itive with 2 of the cases 1196 (0.2%) confirmed by study
criteria [53]. With Texas and Mexico sharing such a large
border region, it stands to reason that the seroprevalence
Fig. 4 Heat map of Chagas disease diagnostic codes and heart-related ICD diagnostics codes combined for 2013–2016
Table 4 County-level descriptive statistics for linear model of possible missed diagnosis of Chagas disease in Texas
Mean SD Median Minimum Maximum
Total 347.575 1157.378 76.500 1.000 12,983.000
Mean age 17.114 0.876 17.098 12.000 23.000
Male 165.278 540.208 39.500 0.000 6158.000
Female 123.310 408.708 25.500 0.000 4597.000
Hispanic 64.060 243.830 7.000 0.000 1944.000
Non-Hispanic 245.730 860.698 53.000 0.000 9888.000
Asian/Pacific Islander 1.274 6.847 0.000 0.000 80.000
Black 17.571 97.493 1.000 0.000 1217.000
White 69.944 194.634 16.000 0.000 1801.000
Other 13.202 60.067 1.000 0.000 787.000
Population 115,054.536 410,461.235 19,211.000 272.000 4,713,325.000
PopKM2 47.109 137.891 8.602 0.107 1166.873
Rate 394.257 159.947 385.607 35.868 1166.667
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Page 9 of 11
Pachecoetal. BMC Infectious Diseases (2022) 22:590
in places such as Texas could be higher, however with
limited surveillance and screening it is hard to know the
true amount of CD in Texas or the U.S.
Limitations
is research is one of a few examining CD through hos-
pital records [53]. is is the first to examine statewide
hospital records in order to qualify the potential for
missed CD diagnosis in Texas. However, this research
focused on potentially-missed diagnosed cases of chronic
Chagas, rather than including acute and indeterminate
chronic forms of CD. Moreover, in examining chronic
CD, the scope of this research was limited to CCC, rather
than looking at other sequelae (i.e., gastrointestinal com-
plications). Furthermore, establishing the criteria for
missed diagnoses of CD was the greatest challenge, given
the lack of research to inform specific risk factors that
account for CCC. us, the risk of misclassification is a
concern.
Inpatient records were exclusively used, rather than
including outpatient records given that CD patients do
not necessarily require a hospitalization to be diagnosed.
CD patients may be unaware of their CD status since the
disease is asymptomatic. Because the patients are asymp-
tomatic, they may be undiagnosed and thus not receiv-
ing appropriate care. Finally, the IPUDF data set does not
provide unique patients, rather enumerates the records.
Ultimately, this highlights the under-estimation of tur
missed diagnoses of CD in Texas.
ICD-9 and ICD-10 heart-related and CD diagnostic
codes were not completely comparable given differences
in their definitions. While ICD-10 denotes the disease
progression (i.e., acute, or chronic), there is no code spec-
ifying the indeterminate form of CD. In ICD-9 there is a
code (086.2) that alludes to the asymptomatic, indetermi-
nate form (i.e., Chagas without mention or organ involve-
ment). Similarly, among the heart-related diagnostics,
there is a cardiomyopathy, excluding Chagas in ICD-9
code but not one for ICD-10. Between 2013 and August
of 2015, a total of 21 records indicated Chagas without
mention of organ involvement. Furthermore, ICD codes
are intended for medical billing and are not confirmed
CD diagnosis. e identified barriers [55] to screening
and diagnosing acute and chronic CD are documented
further highlight the challenges in fully estimating the
true prevalence of CD across the state. Our study utilized
the hospital inpatient data to focus and target educa-
tional efforts throughout Texas.
Recommendations
Future research can further explore the patterns of
missed diagnoses within specific geographical targets.
For example, examining and comparing urban and rural
counties only in contrast to examining patterns through-
out the state; or by examining differences in census tracts
or zip codes). Furthermore, the case definitions for the
missed CD diagnostic codes could be re-evaluated. For
example, additional geospatial and statistical analyses can
be performed on specific counties using only idiopathic
cardiomyopathy diagnoses and comparing to other codes
that accounted for the large number of potential CD
heart-related diagnoses (e.g., other ischemic heart dis-
ease, ischemic cardiomyopathy, unspecified cardiomyo-
pathy). Finally, additional research can map the county
demographics and more specific risk factors for CCC
(i.e., by narrowing the age group).
Secondly, the findings support the need for surveil-
lance systems. An entomological surveillance system
should include the study of natural infection in vectors.
In turn, such systems could facilitate screening for indi-
viduals in communities with documented infestation.
In the human population, a surveillance system would
facilitate an increase in the accuracy, validity, and gener-
alizability of a geospatial analysis. at is, maps that are
created to illustrate the magnitude of CD cases in Texas
would greatly benefit from epidemiological data that is
specific to CD, rather than relying on administrative data
such as the Texas PUDF. For example, recent findings
have helped focus educational approaches, particularly
using an Extension for Community Healthcare Outcomes
(ECHO) model among community healthcare workers
[56] in Texas.
Acknowledgements
Not applicable.
Author contributions
GP: Conceptualization, Investigation, Data Collection, Data Analysis, Writing-
Original Draft; LF: Formal Analysis; JB: Writing- Review and Editing; RS: Writing-
Review and Editing; PSG: Conceptualization, Writing- Review and Editing. The
authors approve the submitted version and confirm their contributions. All
authors read and approved the final manuscript.
Funding
The author(s) received no financial support for the research, authorship, and/
or publication of this article.
Availability of data and materials
The modeling data that generated during the current study are available
online: https:// rpubs. com/R- Minat or/ chaga s1. The coded ICD-9 and ICD-10
hospital data are not publicly available from the corresponding author on
reasonable request. Source data (inpatient public use data file, through 2015)
is available for direct download from the Texas Department of State Health
Services: https:// www. dshs. texas. gov/ thcic/ hospi tals/ Inpat ientp udf. shtm. An
official request must be submitted to request Year 2016.
Declarations
Ethics approval and consent to participate
All methods were carried out in accordance with relevant guidelines and
regulations. The research protocol was reviewed by the Committee for the
Protection of Human Subjects (CPHS) at the University of Texas Health Science
Center (UTHealth) and approved on February 5, 2018 (HSC-SPH-17-1039). No
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Page 10 of 11
Pachecoetal. BMC Infectious Diseases (2022) 22:590
informed consent was required for this protocol since this study utilized a de-
identified administrative data from the Texas Hospital Inpatient Public Use File
for secondary data analysis.
Consent for publication
Not applicable.
Competing interests
The authors declare that they have no competing interests.
Author details
1 School of Health Administration, Texas State University, San Marcos, TX 78666,
USA. 2 School of Public Health, San Diego State University, San Diego, CA
92182, USA.
Received: 22 February 2022 Accepted: 29 June 2022
References
1. Montgomery SP, et al. Neglected parasitic infections in the United States:
Chagas disease. Am J Trop Med Hyg. 2014;90(5):814–8.
2. Bern C, et al. Trypanosoma cruzi and Chagas’ Disease in the United States.
Clin Microbiol Rev. 2011;24(4):655–81.
3. Bern C. Antitrypanosomal therapy for chronic Chagas’ disease. N Engl J
Med. 2011;364(26):2527–34.
4. Bern C et al. Chagas disease in the United States: a public health
approach. Clin Microbiol Rev. 2019; 33(1).
5. Lidani KCF, et al. Chagas disease: from discovery to a worldwide health
problem. Front Public Health. 2019;7:166.
6. Centers for Disease Control and Prevention. Chagas Disease Detailed FAQs.
2013 July 19 [cited 2016 January 1]; Available from: http:// www. cdc. gov/
paras ites/ chagas/ gen_ info/ detai led. html.
7. Pan American Health Organization. General information: Chagas disease.
2016 [cited 2017 August 17]; Available from: http:// www. paho. org/ hq/
index. php? option= com_ conte nt& view= artic le& id= 5856& Itemid=
41506 & lang= en.
8. Manne-Goehler J, et al. Estimating the Burden of Chagas Disease in the
United States. PLoS Negl Trop Dis. 2016;10(11): e0005033.
9. Garcia MN, et al. Knowledge, attitudes, and practices of Texas hunters: a
potentially high-risk population for exposure to the parasite that causes
Chagas disease. Parasit Vectors. 2015;8:197.
10. Rassi A Jr, Rassi A, Marin-Neto JA. Chagas disease. Lancet.
2010;375(9723):1388–402.
11. Appendix 2. Transfusion, 2009. 49(233S). Retrieved from https:// www.
aabb. org/ tm/ eid/ Docum ents/ 231s. pdf.
12. Garcia MN, et al. Historical perspectives on the epidemiology of human
Chagas disease in Texas and recommendations for enhanced under-
standing of clinical Chagas disease in the Southern United States. PLoS
Negl Trop Dis. 2015;9(11): e0003981.
13. Texas Department of Health Services. Chagas disease update- Texas,
2015. 2015; Available from: http:// www. wcchd. org/ servi ces/ docs/ DSHS_
Chagas_ Disea se_ Commu nique_ 09_ 15. pdf.
14. Torpy JM, Burke AE, Glass RM. JAMA patient page. Chagas disease. JAMA.
2007;298(18):2222.
15. Guide for surveillance, prevention, control and clinical management of
acute foodborne Chagas disease. 2009, PANAFTOSA-VP, PAHO, WHO
16. Bern C, et al. Evaluation and treatment of chagas disease in the United
States: a systematic review. JAMA. 2007;298(18):2171–81.
17. Torrico F, et al. New regimens of benznidazole monotherapy and in
combination with fosravuconazole for treatment of Chagas disease
(BENDITA): a phase 2, double-blind, randomised trial. Lancet Infect Dis.
2021;21(8):1129–40.
18. Morillo CA, et al. Randomized trial of benznidazole for chronic Chagas’
cardiomyopathy. N Engl J Med. 2015;373(14):1295–306.
19. Mitelman JE. Consensus statement on Chagas-Mazza disease. Rev Argent
Cardiol. 2011;79(6):544–64.
20. Bonney KM, Engman DM. Chagas heart disease pathogenesis: one
mechanism or many? Curr Mol Med. 2008;8(6):510–8.
21. Kierszenbaum F. Chagas’ disease and the autoimmunity hypothesis. Clin
Microbiol Rev. 1999;12(2):210–23.
22. Viotti R, et al. Towards a paradigm shift in the treatment of chronic Cha-
gas disease. Antimicrob Agents Chemother. 2014;58(2):635–9.
23. Hotez PJ, et al. Chagas disease: “the new HIV/AIDS of the Americas.” PLoS
Negl Trop Dis. 2012;6(5): e1498.
24. Marin-Neto JA, et al. Pathogenesis of chronic Chagas heart disease. Circu-
lation. 2007;115(9):1109–23.
25. Hyson P, et al. Experiences with diagnosis and treatment of Chagas dis-
ease at a United States teaching hospital-clinical features of patients with
positive screening serologic testing. Trop Med Infect Dis. 2021;6(2):93.
26. Bern C, Montgomery SP. An estimate of the burden of Chagas disease in
the United States. Clin Infect Dis. 2009;49(5):e52–4.
27. Forsyth CJ, et al. Current gaps and needs for increasing access to health-
care for people with Chagas disease in the USA. Curr Trop Med Rep.
2019;6(1):13–22.
28. Texas Department of Health Services. Chagas. 2015 [cited 2015 Decem-
ber 1]; Available from: https:// www. dshs. state. tx. us/ idcu/ disea se/ chagas/.
29. Garcia MN, et al. Evidence of autochthonous Chagas disease in south-
eastern Texas. Am J Trop Med Hyg. 2015;92(2):325–30.
30. Gunter SM, et al. Likely autochthonous transmission of Trypanosoma cruzi
to Humans, South Central Texas, USA. Emerg Infect Dis. 2017;23(3):500–3.
31. Cantey PT, et al. The United States Trypanosoma cruzi Infection Study:
evidence for vector-borne transmission of the parasite that causes
Chagas disease among United States blood donors. Transfusion.
2012;52(9):1922–30.
32. Garcia MN, et al. Development of Chagas cardiac manifestations among
Texas blood donors. Am J Cardiol. 2015;115(1):113–7.
33. Garcia MN, et al. Trypanosoma cruzi screening in Texas blood donors,
2008–2012. Epidemiol Infect. 2016;144(5):1010–3.
34. Conners EE, et al. A global systematic review of Chagas disease preva-
lence among migrants. Acta Trop. 2016;156:68–78.
35. World Health Organization et al. Sustaining the drive to overcome the
global impact of neglected tropical diseases: second WHO report on
neglected tropical diseases. 2013, Geneva, Switzerland: World Health
Organization. xii, 138 p.
36. Stigler Granados P, et al. Assessing the effectiveness of Chagas disease
education for healthcare providers in the United States. BMC Infect Dis.
2020;20(743):1–6.
37. Pacheco GJ, Stigler Granados P, Betancourt J. Shaping recommendations
and healthcare provider education on Chagas disease from a mixed
methods baseline exploratory study. Unpublished.
38. Mahoney West H, et al. Perceived barriers to Chagas disease screening
among a diverse group of prenatal care providers. PLoS ONE. 2021;16(2):
e0246783.
39. Texas Department of Health Services. Chagas. 2015 [cited 2015 Decem-
ber1]; Available from: https:// www. dshs. state. tx. us/ idcu/ disea se/ chagas/.
40. Texas Department of State Health Services. Texas Inpatient Public Use
Data File (PUDF). 2020 [cited 2020 June 26]; Available from: https:// www.
dshs. texas. gov/ thcic/ hospi tals/ Inpat ientp udf. shtm.
41. United States Census Bureau. American Fact Finder. 2018 [cited 2018
October 19]; Available from: https:// factfi nder. census. gov/ faces/ nav/ jsf/
pages/ index. xhtml.
42. United States Census Bureau. TIGER/ Line Shapefiles. [cited 2018 March
7]; Available from: https:// www. census. gov/ cgi- bin/ geo/ shape files/ index.
php.
43. 2017 [cited 2017 August 15]; Available from: https:// www. arcgis. com/
featu res/ index. html.
44. How Spatial Autocorrelation (Global Moran’s I) works. [cited 2021 Decem-
ber 17]; Available from: https:// pro. arcgis. com/ en/ pro- app/ latest/ tool-
refer ence/ spati al- stati stics/h- how- spati al- autoc orrel ation- moran-s- i- spati
al- st. htm.
45. Lagrange Multiplier diagnostics for spatial dependence. [cited 2021 Decem-
ber 17]; Available from: https:// rdrr. io/ rforge/ spdep/ man/ lm. LMtes ts. html.
46. Lynn MK, et al. Contemporary autochthonous human Chagas disease in
the USA. Acta Trop. 2020;205: 105361.
47. Sarkar S, et al. Chagas disease risk in Texas. PLoS Negl Trop Dis.
2010;4(10):e836.
48. Forsyth CJ, et al. Recommendations for screening and diagnosis of Cha-
gas disease in the United States. J Infect Dis. 2021;225:1601–10.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Page 11 of 11
Pachecoetal. BMC Infectious Diseases (2022) 22:590
•
fast, convenient online submission
•
thorough peer review by experienced researchers in your field
•
rapid publication on acceptance
•
support for research data, including large and complex data types
•
gold Open Access which fosters wider collaboration and increased citations
maximum visibility for your research: over 100M website views per year
•
At BMC, research is always in progress.
Learn more biomedcentral.com/submissions
Ready to submit your research
Ready to submit your research
? Choose BMC and benefit from:
? Choose BMC and benefit from:
49. Centers for Disease Control and Prevention. What is GIS? 2016 [cited 2018
October 1]; Available from: https:// www. cdc. gov/ gis/ what- is- gis. htm.
50. Webber BJ, et al. A case of Chagas cardiomyopathy following infection in
south central Texas. US Army Med Dep J. 2017;1–17:55–9.
51. Noble D, et al. Feasibility study of geospatial mapping of chronic disease
risk to inform public health commissioning. BMJ Open. 2012;2(1):
e000711.
52. Echavarria NG, et al. Chagas disease: chronic Chagas cardiomyopathy.
Curr Probl Cardiol. 2021;46(3): 100507.
53. Nolan MS, et al. Continuing evidence of Chagas disease along the Texas-
Mexico border. PLoS Negl Trop Dis. 2018;12(11): e0006899.
54. Lima NA, et al. Hospitalization for Chagas heart disease in the United
States from 2002 to 2017. JAMA Netw Open. 2021;4(10): e2129959.
55. Hochberg NS et al. Chagas Disease in the United States: a perspective on
diagnostic testing limitations and next steps. Am J Trop Med Hyg. 2021.
56. Stigler Granados P et al. Assessing healthcare providers’ knowledge and
awareness of Chagas disease utilizing the extension for community
healthcare outcomes model. Unpublished.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in pub-
lished maps and institutional affiliations.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com