Validation of the spanish version of the test for respiratory and asthma control in kids (track) in a population of hispanic preschoolers

Validation of the Spanish version of the Test forRespiratory and Asthma Control in Kids (TRACK) in apopulation of Hispanic preschoolers Carlos E. Rodríguez-Martínez, MD, Gustavo Nino, MD, and Jose A. Castro-Rodriguez, MD, Bogota,Colombia; Washington, DC; and Santiago, Chile What is already known about this topic? The only questionnaire intended to measure asthma control in childrenyounger than 5 years old, the Test for Respiratory and Asthma Control in Kids, needs additional validation studies.
What does this article add to our knowledge? The Spanish version of the Test for Respiratory and Asthma Control inKids questionnaire has excellent content validity, sensitivity to change, and usability; adequate criterion validity, constructvalidity, and test-retest reliability; and an acceptable internal consistency when used in preschoolers with asthma.
How does this study impact current management guidelines? Clinicians who work with Spanish-speakingcommunities have a validated, easy-to-use, parent- and/or caregiver-completed questionnaire with which to assessrespiratory control in children younger than age 5 years with symptoms consistent with asthma.
BACKGROUND: There is a critical need for validation studies internal consistency reliability, and usability of the TRACK of questionnaires designed to assess the level of control of asthma in children younger than 5 years old.
RESULTS: Median (interquartile range) of the TRACK scores OBJECTIVE: To validate the Spanish version of the Test for were significantly different between patients with well-controlled Respiratory and Asthma Control in Kids (TRACK) asthma, patients with not well-controlled asthma, and patients questionnaire in children younger than age 5 years with with very poorly controlled asthma (90.0 [75.0-95.0], 75.0 [55.0- symptoms consistent with asthma.
85.0], and 35.0 [25.0-55.0], respectively, P < .001). TRACK METHODS: In a prospective cohort validation study, parents scores were significantly different between patients classified as and/or caregivers of children younger than age 5 years and with currently symptomatic and symptomatic in the recent past (42.5 symptoms consistent with asthma, during a baseline and a [25.0-55.0] vs 85.0 [75.0-90.0]; P < .001). The intraclass follow-up visit 2 to 6 weeks later, completed the information correlation coefficient of the measurements was 0.755 (95% CI, required to assess the content validity, criterion validity, 0.503-1.00). All patients whose clinical status changed showed construct validity, test-retest reliability, sensitivity to change, an increase of 10 or more points in TRACK score betweenbaseline and follow-up visits. The Cronbach a was 0.77 for thequestionnaire as a whole.
CONCLUSION: The Spanish version of the TRACK questionnaire has excellent sensitivity to change and usability; Department of Pediatrics, School of Medicine, Universidad Nacional de Colombia, adequate criterion validity, construct validity, and test-retest bResearch Unit, Military Hospital of Colombia, Bogota, Colombia reliability; and an acceptable internal consistency, when used in cDivision of Pediatric Pulmonary, Sleep Medicine and Integrative Systems Biology.
children younger than age 5 years with symptoms consistent Center for Genetic Research, Children's National Medical Center, George with asthma.
Ó 2014 American Academy of Allergy, Asthma & Washington University, Washington, DC dDepartments of Pediatrics and Family Medicine, School of Medicine, Pontificia Immunology (J Allergy Clin Immunol Pract 2014;-:---) Universidad Catolica de Chile, Santiago, Chile This work was supported in part by the National Institutes of Health Career Key words: Asthma; Asthma control; Preschool children; Risk; Development Award K12HL090020 and K12HD001399-13, Bethesda, Maryland Validation studies; Reliability; Validity Conflicts of interest: J. A. Castro-Rodriguez has received consultancy fees from The prevalence of asthma has increased since 2001 in the GlaxoSmithKline; has received research support from Novartis; and has received United Statesand is a major public health problem in many lecture fees from GlaxoSmithKline, Novartis, AstraZeneca, and MSD. The rest of countries, for example, Colombia, which has a prevalence esti- the authors declare that they have no relevant conflicts of interest.
Received for publication August 27, 2013; revised November 27, 2013; accepted for mated at 10% to 12%.Over the past decade, the concept of publication January 22, 2014.
asthma control as the degree to which manifestations of the Corresponding author: Carlos E. Rodríguez Martínez, MD, MSc, Avenida Calle 127 disease are reduced or removed by therapy has been clearly No. 20-78. Bogotá, Colombia. E-mail: .
defined and has been incorporated into current asthma guidelines 2213-2198/$36.00Ó 2014 American Academy of Allergy, Asthma & Immunology as a therapeutic goal. Childhood asthma causes considerable morbidity, interference with normal daily activities, and a burden RODRÍGUEZ-MARTÍNEZ ET AL J ALLERGY CLIN IMMUNOL PRACT younger than age 5 years with either physician-diagnosed asthma Abbreviations used or recurrent respiratory episodes suggestive of asthma.
EPR-3- Expert Panel Report-3 IQR- Interquartilic range NAEPP- National Asthma Education and Prevention ProgramTRACK- Test for Respiratory and Asthma Control in Kids The study was undertaken in The Fundacion Hospital La Misericordia, a tertiary care university-based children's hospitallocated in the metropolitan area of Bogota, Colombia. Parents ofchildren younger than 5 years old evaluated in our pediatric for the whole family. It is a major reason for absences from respiratory service from January 2012 to July 2013 with a history school, admissions to the hospital, and visits to the emergency of a child having been diagnosed with asthma or with symptoms department, especially during periods when the asthma is inad- consistent with asthma were invited to participate in the study.
equately controlleIdentification of asthma is particularly Parents of participating children were native Spanish speakers, important in infants and preschoolers because almost 80% of with widely varied educational backgrounds (at least 5 years of patients with asthma start having symptoms during the first 5 formal education) and socioeconomic status but with an years of their life.Moreover, compared with older children with acceptable reading speed and ability. Symptoms consistent with this condition, young children with asthma experience less asthma were operationally defined as either 2 or more episodes of favorable responses to asthma treatments, thus increasing the wheezing, shortness of breath, cough that lasted more than 24 health care utilization and overall burden of the disease.Also, in hours, or use of aerosolized bronchodilator for respiratory developed countries, among all the pediatric population, children symptoms. Children with respiratory conditions not consistent younger than 3 years old have the poorest asthma contr with asthma that might affect the cardiopulmonary status (eg, Several asthma symptom questionnaires that combine indi- chronic lung disease or congenital heart disease) and those with vidual variables to generate a composite score have been devel- other significant chronic disorders or congenital abnormalities oped to measure asthma control in These asthma were excluded from the study.
symptom questionnaires typically assess patient-reported ele-ments of impairment. Some of them also include information TRACK questionnaire about the history of exacerbations and pulmonary function The TRACK questionnaire is a 5-item caregiver-completed parameters.Most of these pediatric asthma control question- questionnaire useful in assessing and monitoring respiratory naires have the shortcoming that they only assess the impairment symptoms in children younger than age 5 years with either domain and not the risk of exacerbation domain as recom- physician-diagnosed asthma or recurrent respiratory episodes mended in the Expert Panel Report-3 (EPR-3) of the National suggestive of asthma. This tool is particularly appealing for Asthma Education and Prevention Program and in asthma care because it assesses both the risk and the impairment the Global Initiative for Asthma (the number of domains.The first 4 items of the TRACK questionnaire exacerbations that requires oral corticosteroids during the previ- evaluate the impairment domain (frequency of respiratory ous 12 months).
symptoms, activity limitation, nighttime awakenings in the past To the best of our knowledge, at present the only question- 4 weeks, and rescue medication use in the past 3 months). The naire intended to measure asthma control in children younger fifth item of TRACK assesses the risk domain (oral corticosteroid than 5 years old that has been developed and validated is the Test use in the previous year) The score for the response to for Respiratory and Asthma Control in Kids (TRACK). The each item ranges from 0 to 20 points, and the scores for the TRACK is a 5-item caregiver-completed questionnaire that individual items are added to obtain the unweighted TRACK assesses both the risk and impairment domains in children questionnaire score. The possible total score ranges from 0 to younger than 5 years of age with either physician-diagnosed 100, with a higher score that indicates better respiratory control asthma or recurrent respiratory episodes suggestive of asthma.
and a score less than 80 indicates poor control.The TRACK Although the TRACK questionnaire has been validated for use in questionnaire was developed by AstraZeneca (London, UK), young the recent Asthma Outcomes Workshop rec- with the assistance of QualityMetric (Lincoln, RI), and is a ommended that additional validation data be gathered before the trademark of AstraZeneca. To perform the present study, TRACK questionnaire can be recommended as a core or sup- AstraZeneca provided the Spanish version of the TRACK plemental measure of asthma Accordingly, there is a questionnaire and granted us permission to use and validate it.
critical need for additional validation studies performed in The questionnaire was forward translated from the original En- different population subgroups (eg, race or ethnicity, socioeco- glish version into Spanish, but no back translations were done.
nomic status, health literacy) and in more languages. In this Before using this version of the TRACK questionnaire, we context, validating the Spanish version of the questionnaire is changed the brand names of the quick-relief medications (item 4) crucial because, according to the last phase III International and oral corticosteroids (item 5) to brand names available in our Study of Asthma and Allergies in Childhood report, 2 of 4 countries with the highest prevalence of asthma in childrenaround the world are Latin American countries where Spanish is Study design and procedures the language (37.6% in Costa Rica and 22.7% in We conducted a prospective cohort validation study that Spanish is one of the most widely spoken languages in the world, followed up a convenience sample of children ages younger than with approximately 500 million native The aim of 5 years old who fulfilled the eligibility criteria (inclusion and the present study was to validate the Spanish version of the exclusion criteria) and whose parents and/or caregivers agreed to TRACK questionnaire in a population of Hispanic children participate and signed the informed consent. All parents and/or J ALLERGY CLIN IMMUNOL PRACT RODRÍGUEZ-MARTÍNEZ ET AL VOLUME -, NUMBER - TABLE I. Spanish version of the TRACK 1. En las últimas 4 semanas, ¿con qué frecuencia le molestaron a tu niño/a los problemas respiratorios, como sibilancias (silbidos), tos o dificultad para 1 vez a la semana 2 ó 3 veces a la semana 4 veces o más a la semana 2. En las últimas 4 semanas, ¿con qué frecuencia despertaron a tu niño/a sus problemas respiratorios (sibilancias (silbidos), tos, dificultad para respirar)? 1 vez a la semana 2 ó 3 veces a la semana 4 veces o más a la semana 3. En las últimas 4 semanas, ¿hasta qué punto interfirieron los problemas respiratorios de tu niño/a, como sibilancias (silbido), tos y dificultad para respirar, con su habilidad para jugar, ir al colegio o participar en las actividades cotidianas que un niño/a debería hacer a su edad?Nunca 4. En los últimos 3 meses, ¿con qué frecuencia tuviste que tratar los problemas respiratorios de tu niño/a (sibilancias (silbidos), tos, dificultad para respirar) con medicinas de alivio rápido (salbutamol, Ventilan, Salbutan, Seramar, Combivent, Berodual)?Nunca 1 vez a la semana 2 ó 3 veces a la semana 4 veces o más a la semana 5. En los últimos 12 meses, ¿con qué frecuencia necesitó tu niño/a tomar corticoesteroides orales (prednisona, prednisolona, Fisopred, Pepred, Medrol, Meticorten) para sus problemas respiratorios no controlados con otras medicinas?Nunca caregivers had an initial visit (baseline), and some of them were symptomatic in the recent past group with a baseline visit that scheduled for a follow-up visit 2 to 6 weeks later. At baseline, we resulted in no change or a step-down in therapy, completed the used standardized forms to collect demographic data of children TRACK questionnaire a second time (2-4 weeks later) to eval- (age, sex) and their respective parents and/or caregivers (age, uate test-retest reliability.
highest level of education), and assessed all the children by usingthe TRACK questionnaire. In addition, at baseline, separately Assessment of the psychometric characteristics of and blinded to the caregiver questionnaire responses, we the TRACK questionnaire collected the timing of respiratory symptoms, the level of asthma To evaluate the TRACK's criterion validity (ie, the degree to control based on NAEPP EPR-3 guidelines recommendations, which the measurement correlates with some other measure of and whether the baseline visit resulted in a step-up in therapy, no the specific construct of control of asthma, such as another change, or a step-down in therapy. The timing of respiratory validated severity instrument or another criterion standard for the symptoms was divided into 3 categories based on the presence or control of asthma), we compared TRACK scores at baseline absence of these symptoms in the previous 4 weeks and in the across the 3 categories of the NAEPP EPR-3 guidelines criteria of previous 12 months as currently symptomatic (episodes of asthma control (poorly controlled, not well-controlled, and well- wheezing, shortness of breath, or coughing in the past 4 weeks), controlled asthma). To assess the TRACK's construct validity (ie, symptomatic in the recent past (episodes of wheezing, shortness the degree to which the measurement corresponds to other of breath, or coughing in the past 12 months but not within the variables and measures not identical but related to the construct past 4 weeks), or asymptomatic (without symptoms for more of control of asthma), we compared baseline TRACK scores than 12 months).
across the 3 categories of the timing of respiratory symptoms To assess the level of asthma control based on NAEPP EPR-3 (currently symptomatic, symptomatic in the recent past, and guidelines recommendations from 0 to 4 years of we asymptomatic), and across the 3 categories of therapeutic deci- completed 5 specific questions that assessed the frequency of sion (a step-up in therapy, no change, or a step-down in therapy).
wheeze, nighttime awakenings from respiratory symptoms, To evaluate the TRACK's test-retest reliability (ie, the consis- activity limitation caused by respiratory symptoms, use of rescue tency of the instrument results measured on 2 occasions with no medications for respiratory symptoms, and oral corticosteroid use change in asthma control in between), we compared TRACK in the previous year. Each question was scored on a 3-point scores in patients classified as symptomatic in the recent past or Likert-type scale (1, well controlled; 2, not well controlled; 3, asymptomatic at baseline in whom no change or a step-down in very poorly controlled). The scoring of these questions was used therapy occurred and who were classified in the same manner in to stratify the sample into categories of very poorly controlled the follow-up visit 2 to 4 weeks later. To assess the TRACK's asthma (if a score of 3 was selected for any question), not well- sensitivity to change (ie, the ability of a score to detect a clinically controlled asthma (if a score of 2 was selected for 1 or more important change over time), we compared TRACK scores in questions and a score of 3 was not selected for any question), or patients classified as currently symptomatic at baseline, in whom well-controlled asthma (if a score of 1 was selected for all 5 the baseline visit resulted in a step-up in therapy, and who were questions). Fifty-eight parents and/or caregivers were selected classified as symptomatic in the recent past in the follow-up visit randomly from the currently symptomatic group at baseline and 4 to 6 weeks later. To examine the TRACK's internal consistency from those with stepped-up therapy after the baseline visit to reliability (ie, the degree of correlation between a scale's items), complete the TRACK questionnaire again 4 to 6 weeks after the we used the responses given for all the parents and/or caregivers initial assessment to evaluate the sensitivity to change. Likewise, at baseline. To assess the TRACK's usability (ie, the speed, 69 parents and/or caregivers selected randomly from the understandability, and subjective experience when completing RODRÍGUEZ-MARTÍNEZ ET AL J ALLERGY CLIN IMMUNOL PRACT the questionnaire), parents and/or caregivers were asked to NAEPP EPR-3 guidelines recommendations was well-controlled qualify the ease of scoring of the TRACK questionnaire as easy to asthma in 82 children (34.2%), not well-controlled asthma in score, moderately easy to score, or difficult to score. In addition, 52 (21.7%), and very poorly controlled asthma in 106 (44.2%).
the time to complete the questionnaire was reported. The study In relation to the therapeutic decision, in 70 patients (29.2%), protocol was approved by the local ethics board.
the baseline visit resulted in no change in therapy, in116 (48.3%) in a step-up in therapy, and in 54 (22.5%) in a Statistical analysis step-down in therapy. The median (interquartile range [IQR]) of To assess the TRACK's criterion validity, we compared the TRACK scores of the 240 patients included in the study TRACK scores across the 3 categories of the NAEPP EPR-3 was 65.0 (40.0-85.0) points. With respect to the most frequent guidelines criteria of asthma control (poorly controlled, not well- responses for each item, 100 (41.7%) of the parents and/or controlled, and well-controlled asthma) by using 1-way ANOVA caregivers responded "not at all" to the first item, 124 (51.7%) or the Kruskall-Wallis nonparametric method, as appropriate. To responded "not at all" to the second item, 118 (49.2%) assess the TRACK's construct validity, we contrasted TRACK responded "not at all" to the third item, 84 (34.2%) responded scores across the 3 categories of the timing of respiratory "once or twice" to the fourth item, and 56 (23.3%) responded symptoms (currently symptomatic, symptomatic in the recent "twice" for the fifth item.
past, and asymptomatic) and across 2 categories of therapeuticdecision (a step-up in therapy vs no change or a step-down intherapy) by using the 1-way ANOVA or the Kruskall-Wallis test, Criterion validity as appropriate. Test-retest reliability was assessed with the TRACK scores were significantly different among patients intraclass correlation coefficient and the Lin concordance corre- with well-controlled asthma, patients with not well-controlled lation coeffiand through the construction of the Bland asthma, and patients with very poorly controlled asthma (median and Altman plot.The TRACK's sensitivity to change was [IQR] 90.0 [75.0-95.0], 75.0 [55.0-85.0], and 35.0 [25.0-55.0], determined by comparing TRACK scores at baseline and follow- respectively; P < .001) (see in this article's Online up with the paired Student t test or the Wilcoxon signed rank Repository at ).
test, as appropriate, and by calculating the proportion of patientswith an increase of 10 or more points in TRACK scores between Construct validity baseline and follow-up measurements because this value repre- TRACK scores were significantly different between patients sents a clinically meaningful change in respiratory control status classified as currently symptomatic and symptomatic in the when using the TRACK questionnaire.Internal consistency recent past (median [IQR] 42.5 [25.0-55.0] vs 85.0 [75.0-90.0]; reliability was assessed by using the Cronbach a coefficient.
P < .001] (see in this article's Online Repository at The use of the method proposed by Walter et alto calculate ). Likewise, TRACK scores were signifi- the required number of subjects in a reliability study, in which cantly different between patients whose baseline visit resulted in reliability is measured, yielded a sample size of 64 patients, 2 a step-up in therapy compared with patients for whom this visit methods to be reported in the diagnosis, a kappa for the null resulted in no change or a step-down in therapy (40.0 [25.0- hypothesis of 0.5, a kappa for the alternative hypothesis of 0.7, a 55.0] vs 85.0 [75.0-90.0]; P < .001) (see in this statistical significance level of .05, and a power of 80%. In the article's Online Repository at assessment of criterion and construct validity, 64 participantsprovide at least 80% power (effect size, 0.4; a ¼ 0.05; 2-sided Test-retest reliability test) to detect differences among the groups by using 1-way The median (IQR) of the TRACK scores at baseline and in ANOVA.Statistical analysis was done with Stata 12.0 (Stata the follow-up visits were not significantly different (85.0 [76.2- Corp, College Station, Tex).
90.0] vs 90.0 [80.0-90.0]; P ¼ .11). The intraclass correlationcoefficient of the measurements was 0.755 (95% CI, 0.503- 1.00), and the Lin concordance correlation coefficient was 0.814 Patient population (95% CI, 0.733-0.894). The Bland and Altman plot shows the Of the total number of patients who fulfilled the eligibility agreement of TRACK scores between baseline and follow-up criteria (n ¼ 242), 2 were excluded because their parents refused visits. The mean difference in the TRACK score between the 2 to participate in the study, so 240 (99.2%) were finally enrolled visits was 2.50, and their corresponding 95% limit of agree- in the study. Likewise, 2 of the total of 52 patients (3.8%) ment was 12.53 to 7.53 (). Two outliers were found, eligible for sensitivity to change assessment and 1 of the total of and the points in the plot show a random distribution.
65 patients (1.5%) eligible for test-retest reliability evaluation didnot attend the follow-up visit, so, finally, 50 and 64 patients were Sensitivity to change included in these groups, respectively. The mean (SD) of the ageof the 240 patients included in the study was 40.02  10.9 TRACK scores at baseline were significantly lower than those months. The age group distribution was as follows: 17 (7.1%) obtained in the follow-up visit (median [IQR] 35.0 [10.0-50.0] < 24 months, 167 (69.6%) between 25 and 48 months, and the vs 72.5 [65.0-80.0]; P < .001). All the patients showed an in- remaining 56 (23.3%) between 49 and 60 months. Of the total crease of 10 or more points in TRACK score between baseline of patients, 136 patients (56.7%) were men. With respect to the and follow-up visits.
timing of respiratory symptoms at baseline, 124 (51.7%) wereclassified as currently symptomatic and 116 (48.3%) as symp- Internal consistency tomatic in the recent past. No patient was classified as asymp- The Cronbach a was 0.77 for the questionnaire as a whole.
tomatic. At baseline, the level of asthma control based on For the individual items, this statistic ranged from 0.65 to 0.85.

J ALLERGY CLIN IMMUNOL PRACT RODRÍGUEZ-MARTÍNEZ ET AL VOLUME -, NUMBER - FIGURE 1. Bland and Altman plot, displaying the difference in TRACK scores plotted against the mean TRACK scores. Horizontal lines are drawn at the mean difference and at the mean difference 1.96 SD of the differences. In the figure, certain data points are falling on topof each other.
which may lead to significant reductions of the disease burden All parents qualified the TRACK as easy to score, and the time required to complete the questionnaire ranged from 45 to 60 Our results with respect to criterion and construct validity of the TRACK are consistent with those reported by Murphyet al,who found that mean scores of the TRACKdiffered significantly in the expected direction for 3 levels of NAEPP EPR-3 guidelinesebased respiratory control, physician- The present study showed that the Spanish version of the recommended change in therapy, and symptom status. Similarly, TRACK questionnaire has adequate psychometric characteristics Chipps et also found significant differences in mean TRACK when tested in children younger than age 5 years with either scores among children categorized according to the physician's physician-diagnosed asthma or recurrent respiratory episodes NAEPP EPR-3 guidelinesebased control table ratings. In suggestive of asthma as demonstrated by (1) an adequate crite- agreement with our results, they also found that children who rion validity when we compared TRACK scores across the 3 received a recommendation for a step-up in therapy from their categories of the NAEPP EPR-3 guidelines criteria of asthma physician scored significantly lower on TRACK at baseline control, (2) an adequate construct validity when we compared relative to children who received a recommendation for main- TRACK scores across the 3 categories of the timing of respiratory tained or stepped-down therapy. With respect to TRACK's test- symptoms and across 2 categories of therapeutic decision, (3) an retest reliability, the point estimate value of the intraclass cor- acceptable internal consistency, and (4) an excellent usability relation coefficient that we found was higher than that reported when parents and/or caregivers evaluated its ease of scoring and in the study by Chipps et al,although the 95% CI of our the time to complete the questionnaire. Likewise, the Spanish estimation contains the value found in their study. This higher version of the TRACK questionnaire showed adequate test-retest value in our study is probably because our time period between reliability and excellent sensitivity to change when we compared the baseline and follow-up was shorter than that used in the baseline and follow-up scores in Hispanic preschoolers with study by Chipps et thus increasing the likelihood of a major asthma symptoms.
consistency of the questionnaire results between the baseline and The findings of this study are important because they will follow-up visits.
encourage physicians who work with Spanish-speaking com- Regarding the TRACK's sensitivity to change, our findings are munities to use the TRACK questionnaire for clinical and consistent with those reported by Chipps et who found research purposes. Implementing the use of the Spanish version significant differences in TRACK scores across patient subsets of the TRACK questionnaire for assessing asthma control that differed in change status (better, same, or worse) on the basis in Hispanic preschoolers with this condition may allow the of the physician's guidelines-based control table ratings and NAEPP EPR-3 recommended stepwise approach for childhood caregivers' reports. In terms of TRACK's internal consistency, in asthma management (step-up if necessary and step-down when agreement with our findings, the Cronbach a values reported in possible) in Spanish-speaking low-middle income countries (eg, the studies by Murphy et aland Chipps et alwere above the Colombia) and among Latino immigrant families worldwide, 0.7 recommended reliability threshold value for multi-item RODRÍGUEZ-MARTÍNEZ ET AL J ALLERGY CLIN IMMUNOL PRACT scales. However, in the study by Chipps et this value was obtained only after deleting item 5 from the scale, so it may reflect not only the different nature of the domain that assesses item 5 (risk) compared with that assessed by items 1 to 4 (impairment) but also the independence in the variation of these 2 domains. To the best of our knowledge, no other studies have 5. American Lung Association Epidemiology and Statistics Unit Research and previously reported the usability parameters of speed, under- Program Services. Trends in asthma morbidity and mortality 2005. Availablefrom: standability and subjective experience that we found in our study . Accessed June 28, 2013.
when parents and/or caregivers completed the TRACK The main limitations of our study comprise the relatively small number of patients included and that the study was per- formed in a single center and in a unique clinical setting (out- patients). Although it is probable that the Spanish version of the TRACK questionnaire exhibits similar psychometric properties in other populations and other clinical settings, it is necessary to conduct additional studies to determine the psychometric char- acteristics of the TRACK questionnaire in a larger number of patients, in different settings, with a more varied educational background, and with a more representative sample of the gen- eral population of preschoolers with symptoms consistent with asthma. The main strength of our study is the assessment of all of the recommended psychometric characteristics in the validation process of severity scores and other outcome measures of the TRACK questionnaire in Spanish, one of the most widely spoken languages in the world. In addition, to the best of our knowledge, a formal assessment of TRACK's usability had not been previously reported.
In summary, analysis of our results suggests that the Spanish 14. Global Strategy for Asthma Management and Prevention, Global Initiative for Asthma (GINA) 2012. Available from: . Accessed version of the TRACK questionnaire has an adequate criterion June 28, 2013.
validity, an adequate construct validity, an adequate test-retest reliability, an excellent sensitivity to change, an acceptable internal consistency, and an excellent usability when used in children younger than age 5 years with either physician-diagnosed asthma or recurrent respiratory episodes suggestive of asthma. Given that 16. Languages of the world: interesting facts about languages. Available from: asthma is highly prevalent among Hispanic children who live in . Accessed August 4, Latin America and in non-Spanish speaking countries (eg, the United States), the validation of the Spanish version of the TRACK questionnaire may lead to significant reductions in health disparities in pediatric asthma care worldwide.
We thank Charlie Barret for his editorial assistance.
J ALLERGY CLIN IMMUNOL PRACT RODRÍGUEZ-MARTÍNEZ ET AL VOLUME -, NUMBER - TABLE E1. Frequency of responses for each item of the TRACK questionnaire at baseline according to the level of asthma control based on Expert Panel Report-3 of the National Asthma Education and Prevention Program (NAEPP EPR-3) guidelines* Well-controlled asthma, no. (%), Not well-controlled asthma no. (%), Very poorly controlled asthma no. (%), Two or 3 times a wk Four or more times a wk Two or 3 times a wk Four or more times a wk Two or 3 times a wk Four or more times a wk Four or more times RODRÍGUEZ-MARTÍNEZ ET AL J ALLERGY CLIN IMMUNOL PRACT TABLE E2. Frequency of responses for each item of the TRACK questionnaire at baseline according to the timing of respiratory Current symptomatic, no. (%), n [ 124 (51.7%) Recent past symptomatic, no. (%), n [ 116 (48.3%) Two or 3 times a wk Four or more times a wk Two or 3 times a wk Four or more times a wk Two or 3 times a wk Four or more times a wk Four or more times J ALLERGY CLIN IMMUNOL PRACT RODRÍGUEZ-MARTÍNEZ ET AL VOLUME -, NUMBER - TABLE E3. Frequency of responses for each item of the TRACK questionnaire at baseline according to the therapeutic decision at No change in therapy, no. (%), Set-up in therapy, no. (%), Step-down in therapy, no. (%), Two or 3 times a wk Four or more times a wk Two or 3 times a wk Four or more times a wk Two or 3 times a wk Four or more times a wk Four or more times

Source: http://www.neumopediatriacolombia.com/wp-content/uploads/2015/07/Validaci%C3%B3n-del-TRACK-en-espa%C3%B1ol1.pdf