ORIGINAL RESEARCH |
https://doi.org/10.5005/jp-journals-10030-1431 |
Community and Internet-driven Development of the Spanish Language Version of the Injured Trauma Survivor Screen
1Comprehensive Injury Center, Medical College of Wisconsin, Milwaukee, Wisconsin, United States of America
2Department of Surgery, Division of Trauma & Acute Care Surgery, Medical College of Wisconsin, Milwaukee, Wisconsin, United States of America; Comprehensive Injury Center, Medical College of Wisconsin, Milwaukee, Wisconsin, United States of America
Corresponding Author: Amber Brandolino, Comprehensive Injury Center, Medical College of Wisconsin, Milwaukee, Wisconsin, United States of America, Phone: +14149557699, e-mail: abrandolino@mcw.edu
Received: 01 July 2023; Accepted: 30 October 2023; Published on: 30 December 2023
ABSTRACT
Aims and background: The goal of this study was to develop a clinically appropriate translation of the Injured Trauma Survivor Screen (ITSS) from English to Spanish that retained the original intent of each item while creating a culturally valid screening tool.
Materials and methods: This was a cross-sectional study of an existing screen developed to detect risk for posttraumatic stress disorder (PTSD) and depression following injury. Participants were recruited from a local clinic (n = 8) and online via Amazon’s MTurk platform (n = 100) and were asked to translate each of the nine items of the ITSS screen from English into Spanish. The most frequent translations for each item were selected for the draft screen. The draft screen was back-translated by five bilingual individuals. Only two wording changes were made to the draft screen due to the back translations.
Results: The forward translation that resulted from the internet sample provided the product that was then back-translated by five predetermined bilingual individuals. Two changes were made to the end product following the back translation.
Conclusion: The Injured Trauma Survivor Screen–Spanish (ITSS-S) constitutes a strong, methodical translation of a clinical screen for PTSD and depression risk. As such, it has preliminary clinical utility and readiness for further validation.
Clinical significance: A Spanish-language version of the ITSS that was methodologically rigorously forward and back-translated can increase access to formalized screening and intervention for PTSD and/or depression. Future work is needed to further understand the validity and reliability of ITSS-S.
RESUMEN
Objetivos y antecedentes: El objetivo de este estudio fue desarrollar de una traducción clínicamente apropiada del Injured Trauma Survivor Screen (ITSS) de inglés al español que conservara la intención original de cada pregunta y al mismo tiempo creara una pantalla culturalmente válida.
Materiales y métodos: Este fue un estudio transversal de una pantalla existente desarrollada para detectar el riesgo de trastorno de estrés postraumático (TEPT) y depresión después de una lesión. Los participantes fueron reclutados en una clínica local (n = 8) y en línea a través de la plataforma MTurk de Amazon (n = 100) y se les pidió que tradujeran cada uno de los 9 elementos de al pantalla ITSS del inglés al español. Se seleccionaron las traducciones mas frecuentes para cada pregunta para la pantalla borrador. El borrador de la pantalla fue traducido por 5 personas bilingües. Solo se hicieron dos cambios de redacción en la pantalla de borrador debido a las traducciones inversas.
Resultados: La traducción directa que resultó de la muestra de Internet proporciono el producto que luego fue traducido inversamente por 5 personas bilingües predeterminadas. Se realizaron doce cambios en el producto final después de la traducción inversa.
Conclusiones: El ITSS-S, o la versión española del ITSS, constituye una traducción sólida y metódica de una prueba clínico por el riesgo de TEPT y depresión. Como tal, tiene utilidad clínica preliminar y está listo para validación adicional.
Importancia clínica: Una versión española de ITSS que fue traducida metodológicamente rigurosamente directa y inversa puede aumentar el acceso a la detección de riesgo e intervención formalizadas para el riesgo de TEPT y/o la depresión. Es necesario trabajar en el futuro para comprender mejor la validez y fiabilidad de ITSS-S.
How to cite this article: Brandolino A, deRoon-Cassini TA. Community and Internet-driven Development of the Spanish Language Version of the Injured Trauma Survivor Screen. Panam J Trauma Crit Care Emerg Surg 2023;12(3):152–161.
Source of support: Nil
Conflict of interest: None
Keywords: Depression, Development study, Injured Trauma Survivor Screen, Posttraumatic stress disorder, Risk screen, Spanish language, Traumatic injury
Palabras Clave: lesión traumática, trastorno de estrés postraumático, depresión, pantalla de riesgo, lengua española, Injured Trauma Survivor Screen, estudio de desarrollo
INTRODUCTION
The racial composition of the population of the United States is increasingly diversifying. The most recent decennial Census revealed that a significant growth of 25% in non-Hispanic Whites drove the increase.1 Within this subset, Hispanic persons of any race had the largest representation. And while nationally, English is still the most spoken language at home (78.4%), the next most common language is Spanish (13.2%). While this may not readily appear significant, of non-English speakers, Spanish speakers represent over 60% of this cohort, with 39% able to speak English ”less than very well.”2
This shifting composition necessitates the reimagining of the breadth and accessibility of standard-of-care procedures and resources in trauma centers across the country. Trauma centers have been recommended to screen for posttraumatic stress disorder (PTSD) and depression, with intervention to follow in those at risk for long-term distress.3 However, as of January 2023, this is now a mandate made by the American College of Surgeons Committee on Trauma (ACS COT).4 Yet, there exist very few screening instruments validated on an injury population, with even fewer available in Spanish.
This contributes to preexisting healthcare disparities for Spanish-speaking patients. Hispanic adults with limited English proficiency receive one-third less health care (i.e., expenditures) than Hispanic adults with English proficiency.5 Within the traumatically injured patient population specifically, less than half received even a complete motor and sensory exam upon presentations, and >25% were asked about the situational factors of their injury.6 This is one example that led to Spanish-speaking trauma patients reporting feeling a lack of autonomy, empowerment, and understanding of their care.7 A ready-to-use risk screen in Spanish would be one-way to begin to remediate this disparity.
It is insufficient and inefficient to rely on medical interpreters to conduct on-the-spot translation of a risk screen used nationally. Not only are interpretation (of spoken word) and translation (of written word) two different skill sets, but they also result in unstandardized wording. Similar to English, there are multiple ways and words that may be employed to convey the same idea in Spanish. To utilize a screening tool in this way invalidates the purpose of creating standardized and psychometrically evaluated measures for clinical use.
In English, the Injured Trauma Survivor Screen (ITSS) has undergone individual and multicenter validation with hospitalized injury patients at level 1 trauma centers.8–10 The ITSS offers adequate sensitivity and specificity in detecting risk for the development of PTSD and depression following injury at 1- and 6-months postinjury. It is a nine-item screen that is administered by a social worker or another medical provider to patients hospitalized due to injury and has been utilized in trauma centers across the United States as a part of standard-of-care screening practices.
As the screen was developed and validated only in the English language to date, the current study aims to enhance the cross-cultural clinical utility of the ITSS. The benefit of a translation of ITSS into Spanish would be the natural extension of current resources to the next largest racial minority group in the United States, representing a heterogeneous mix of fluent bilinguals, limited English proficiency, and Spanish-only speakers. The outcome of the current study is the Injured Trauma Survivor Screen–Spanish (ITSS-S).
MATERIALS AND METHODS
Design
This was a mixed-method, community-based study aimed to generate a Spanish translation of the ITSS in a way that ensures a culturally and linguistically appropriate product that maintains the clinical intention of the original items. Participants involved in any aspect of the study procedure (reviewed below) were at least 18 years of age and had at least a functional knowledge of the English and Spanish languages. The identified community driving the translation are self-identified, English-Spanish bilingual individuals.
Analytic Framework
The overall strategy of this study follows the suggestion published by Regmi et al.,11 which was synthesized from several other sources: (1) determine the relevance or context; (2) forward-translation; (3) backward-translation; (4) compare both the source and target language products; and (5) revisiting the process to get similar interpretations. This study omits the last step as future work will focus on piloting the end product in a separate study.
In order to allow for the most effective coding of the written words of provided translations, word indices (i.e., tabulations of frequencies of different translations) were created for the data for each item of the ITSS. Then, the wording of each item was broken down into the smallest grammatical units of analysis that could retain meaning independently. Described another way, these units were the smallest linguistic units that carried either an ideational (passivation, nominalization), interpersonal (pronoun use), or textual (word order or repetition) function.12
In order to allow for the most flexibility in interpreting the linguistic nature of the data, the analysis portion of the study was performed manually. Despite the decision to manually review coding patterns and tally frequencies, this was approached with strict adherence to pattern coding to allow for analysis to follow the wording of the participants—everything from word choice, plurality, conjugation, prepositions, article use, etc.
The type of coding used to guide the forward translation was first pattern coding and then in vivo coding. Pattern coding refers to the recurring patterns found in the data that are used as the foundation for data analysis.13,14In vivo coding is simply using the wording of the study participants as the coding for data analysis rather than codes created a priori by an investigator.14 This decision regarding coding was ideal for this study because it strengthened the voice and input of the participants over that of the authors and thereby reduced the opportunity for bias. It also enhanced the ability to capture the full range of participant responses. This was then further enhanced with in vivo coding to further restrict data analysis to only the language used by participants. Doing so excludes the opportunity for linguistic interjection on behalf of the investigator.
For back translation, the goal was to understand if the ITSS-S produced from the forward translation had any significant wording differences between its original English version and the back-translated version that could alter the clinical intent of the screen’s items. There was also the opportunity to see which items, if any, were written too colloquially in English to survive translation into another language.
Participants
The goal of the study was to ensure that community voices were lifted during the translation process over a single translation offered by a translation service company. Enrolled participants were from a healthy (non-injured) population. This ”healthy” community sample was obtained through convenience and snowballing sampling at a local, natural health community clinic located close to the level 1 trauma center of an urban, Midwestern city. There was also a ”healthy” internet sample recruited via purposive sampling through Amazon’s Mechanical Turk (MTurk) project platform, where the requirement of bilingualism in Spanish was programmatically enforced. This choice allowed for a greater sample size, convenience sampling, and both a local and international level of participation. The latter point was important given the regional and national dialects of Spanish. More information about the recruitment of participants follows in the next section.
Procedure
The study consisted of two phases: a forward translation of the ITSS and a back translation of the resultant draft screen from the forward translation. For the local community participants involved in the forward translation, a study packet was located in a secured box at a community health agency that serves largely Latino patients. People were alerted to the study by the primary author through presentations of the study’s intent and procedures at various staff meetings.
The internet sample was recruited from a Qualtrics survey link advertised through MTurk. MTurk is an Amazon service that operates like a crowdsourcing internet marketplace.15 It is a place where people, referred to as Workers, can go to perform ”projects” that computers are unable to do on the Amazon-based platform for financial incentives. MTurk workers represent a diverse, on-demand, scalable, and selective workforce that can choose to complete different tasks for which they qualify to complete. As previously mentioned, the requirement of bilingualism (English and Spanish) was programmatically enforced through the project’s Worker Requirements set-up page. This was done by requiring that only workers with the premium qualification of ”language fluency (basic)–Spanish” would have access to complete this project.16,17 While these and other qualifications are self-identified, they are not modifiable after an MTurk worker has set up their account to avoid inappropriate access to complete projects merely for the financial incentive offered for project completion.15 It thus motivates Workers to answer honestly about their qualifications so they may be paired with work that they will be able to complete and thus be able to be paid.
The back translation from Spanish into English was done by five bilingual persons. Bilingual individuals were recruited from persons known to the primary author as being native speakers of Spanish.
Measures
Participants in the forward translation phase of the study received an informational study letter, a demographic information form, and a guided ITSS translation worksheet. Participants in the back translation phase of the study received an informational study letter and a guided ITSS-S (resultant from the prior phase) back translation worksheet. All documents were written in both languages, except the ITSS or ITSS-S as those were the measures being translated. Different font colors were used to make it clear which part of each document was in English and which part was in Spanish. The primary author served as the translator and creator of all study documents.
Demographic Information
Basic demographic information collected was limited to essential information related to this study’s outcome. The information was not Protected Health Information and was non-identifiable. Information collected included age, race/ethnicity, years of education, sex, language ability in both English and Spanish, and if the participant is an immigrant, how many years they have lived in the United States, and their country of origin. Language ability in both languages was asked separately for writing and speaking ability. This helped to elucidate to what degree the participant was bilingual.
Injured Trauma Survivor Screen (ITSS) Worksheet
The ITSS is comprised of nine items and takes about 5 minutes to administer to inpatient trauma patients by a trained clinician or researcher.7 It measures the risk of developing PTSD and depression at 1-month posttraumatic injury. It was normed on the traumatically injured inpatient population. The screen’s items account for pre-, peri-, and posttrauma risk factors. Each item is answered as either a yes or no. A ”yes” scores a 1, and a ”no” scores a 0 toward the overall sum score for PTSD and/or depression risk. Items 1, 2, 3, 5, and 6 measure PTSD risk, and items 3, 4, 7, 8, and 9 measure depression risk. A sum score of ≥2 indicates a positive screen for risk for either diagnosis.
The screen was reconstructed into a worksheet such that in one column, there was a row for each screen item, plus headings and instructions. In the second column immediately adjacent, there were blank cells within which participants could enter their Spanish translation. There are bilingual instructions at the top of the worksheet to orient participants to the worksheet.
Injured Trauma Survivor Screen–Spanish (ITSS-S) v1 Worksheet
The resultant Spanish translation of the ITSS after the first phase of the study was structured into a worksheet exactly similar to the one received by the forward translation participants. The only difference was replacing ”ITSS” with ”ITSS-S v1” in the header.
Data Analysis
Phase 1: Forward Translation
For the quantitative data, appropriate descriptive statistical analyses were performed to characterize the sample. This pertains to the demographic information (e.g., gender and race) and collected language ability in English and Spanish. Frequencies and raw numbers were used for this characterization.
For the qualitative data, a combination of in vivo and pattern coding was utilized in order to extract common units of wording, phrases, and translations from the ITSS worksheet. A content analysis was performed to allow for the quantification of the frequencies of the different ways each item was translated. The codes extracted constituted the minimum unit of language possible within the wording of each item. For example, within a sentence, there will be, at minimum, a noun and a verb. Within that one sentence, or item, there are two codes that will be derived—the noun and the verb. Different nouns may be used to convey the same idea, and different verbs may not only be selected but conjugated differently. These possible points of variation were considered codes and quantified. The most common translation of each unit was selected to constitute the translation of each item. Thus, the end product from the forward translation is the first version of the ITSS-S (v1) and will represent the most common translation from participants.
Phase 2: Backward Translation
The ITSS-S v1 worksheet was analyzed using a combination of in vivo and pattern coding in order to extract common wording, phrases, and translations to allow a greater understanding of potential areas of discrepancy between the forward translation (ITSS-S v1) and the back translations resulting from Phase 2. These back translations helped to inform if any parts of ITSS-S v1 needed to be linguistically modified in order to hold up the clinical integrity of the screen. The result from this backward translation phase of the final version of ITSS-S.
Data Quality
There were two sources of professional medical translation that were leveraged to ensure translations represented authentic attempts to accurately (forward and back) translate the ITSS. First, the first author has 2 decades of Spanish-language experience—formal classroom education, real-life experience, and 5 years of medical translation at a clinic. They were thus able to carefully review all collected translations to ensure that translations represented reasonable and accurate attempts. Had there been any nonsensical translations from MTurk workers motivated by the financial incentive of the work, the first author would have excluded their responses. It should be noted that this situation did not arise, and so there was not the need to remove participants’ translations, and it is thus not discussed in the results that follow. Second, the first author was provided with a forward translation of the ITSS from a medical translation company that had partnered with the local clinic from which the in-person sample was obtained. This translation was reviewed and compared against the community-driven ITSS-S resulting from this study. The company did not provide a back translation; thus, only the forward translation is discussed with the study sample resultant ITSS-S.
RESULTS
Phase 1: Forward Translation
Participants
There was a total of 108 participants. There were eight participants from the clinic (7.4%) and 100 from the internet sample. Most participants were male (55.6%) and self-identified as Latinx (28.7%) (Table 1). There were 12 different self-reported racial identities, including mestizo, Spanish (Spain), Hispanic and White, Black, ”Human,” and Hispanic and Latinx. Clinic-based participants were about 10 years older than the internet sample [mean (M) = 43.4 years, standard deviation (SD) = 14.98 vs M = 34.64, SD = 8.97 years], while the internet sample had more years of education (M = 15.62 years, SD = 2.92 vs M = 12.80 years, SD = 4.87). For those not born in the United States, the clinic-based sample averaged about 23.5 years in the country (SD = 11.39), while many internet sample participants have not spent time in the United States, thus lowering their average to about 9 years (M = 8.89, SD= 11.22). The main countries of origin were Venezuela (21.3%), Mexico (9.3%), and Peru (4.6%).
| Characteristic | Clinic sample (N = 8) | MTurk sample (N = 100) | ||
|---|---|---|---|---|
| n | % | n | % | |
| Sex | ||||
| Female | 2 | 25.0 | 42 | 42.0 |
| Male | 3 | 37.5 | 57 | 57.0 |
| Other (agender) | 0 | 0.0 | 1 | 1.0 |
| Omitted | 3 | 37.5 | 0 | 0.0 |
| Race | ||||
| Hispanic | 3 | 37.5 | 24 | 24.0 |
| Latinx | 1 | 12.5 | 30 | 30.0 |
| White | 0 | 0.0 | 24 | 24.0 |
| Spanish (Spain) | 0 | 0.0 | 3 | 3.0 |
| ”Human” | 0 | 0.0 | 1 | 1.0 |
| Mestizo | 0 | 0.0 | 3 | 3.0 |
| Mexican | 0 | 0.0 | 2 | 2.0 |
| Hispanic and Latin | 0 | 0.0 | 1 | 1.0 |
| Hispanic and White | 0 | 0.0 | 4 | 4.0 |
| Latinx and White | 0 | 0.0 | 5 | 5.0 |
| Hispanic, White, and American | 0 | 0.0 | 2 | 2.0 |
| Black | 0 | 0.0 | 1 | 1.0 |
| Omitted | 4 | 50.0 | 0 | 0.0 |
| English writing proficiency | ||||
| Native | 1 | 12.5 | 29 | 29.0 |
| Writes with occasional errors | 2 | 25.0 | 56 | 56.0 |
| Writes with frequent errors | 0 | 0.0 | 6 | 6.0 |
| Can write basic sentences | 0 | 0.0 | 6 | 6.0 |
| Can write some words/phrases | 0 | 0.0 | 3 | 3.0 |
| Omitted | 5 | 62.5 | 0 | 0.0 |
| English speaking proficiency | ||||
| Native | 0 | 0.0 | 25 | 25.0 |
| Fluent | 1 | 12.5 | 47 | 47.0 |
| Proficient | 0 | 0.0 | 20 | 20.0 |
| Conversational | 1 | 12.5 | 4 | 4.0 |
| Basic knowledge | 0 | 0.0 | 4 | 4.0 |
| Omitted | 6 | 75.0 | 0 | 0.0 |
| Spanish writing proficiency | ||||
| Native | 2 | 25.0 | 90 | 90.0 |
| Writes with occasional errors | 2 | 25.0 | 9 | 9.0 |
| Writes with frequent errors | 0 | 0.0 | 1 | 1.0 |
| Omitted | 4 | 50.0 | 0 | 0.0 |
| Spanish speaking proficiency | ||||
| Native | 2 | 25.0 | 93 | 93.0 |
| Fluent | 1 | 12.5 | 7 | 7.0 |
| Conversational | 5 | 62.5 | 0 | 0.0 |
| Country of origin (if not United States) | ||||
| Mexico | 3 | 100.0 | 7 | 7.0 |
| Venezuela | 0 | 0.0 | 23 | 23.0 |
| Peru | 0 | 0.0 | 5 | 5.0 |
| Puerto Rico | 0 | 0.0 | 3 | 3.0 |
| Spain | 0 | 0.0 | 4 | 4.0 |
| Chile | 0 | 0.0 | 2 | 2.0 |
| Argentina | 0 | 0.0 | 2 | 2.0 |
| Cuba | 0 | 0.0 | 1 | 1.0 |
| Colombia | 0 | 0.0 | 5 | 5.0 |
| Guatemala | 0 | 0.0 | 1 | 1.0 |
| Nicaragua | 0 | 0.0 | 1 | 1.0 |
| Characteristic | M | SD | M | SD |
| Age | 43.40 | 14.98 | 34.64 | 8.97 |
| Years of education | 12.80 | 4.87 | 15.62 | 2.92 |
| Years in the United States (for those not born in the United States) | 23.50 | 11.39 | 8.89 | 11.22 |
The majority of participants from both samples wrote English with occasional errors (n = 58, 53.7%), spoke English fluently (n = 48, 44.4%), were native Spanish speakers (n = 92, 85.2%), and wrote Spanish with native fluency (n = 95, 88.0%).
Translation
Translation decisions based on frequency non-majorities and unclear majorities, as well as sub-analyses, will be discussed. A non-majority decision was made right away in the title. For the phrase ”injured trauma,” the final translation was lesión traumática. Though lesión did have the majority (53.00%), traumática did not (34.00%). In a subanalysis of the word order for ”injured” and ”trauma,” frequencies for the use of these words as a noun vs adjective were quantified for participants who did not omit either word in their translation. The most common usage was ”injury” as a noun and ”trauma” as an adjective (45.12%). As both words have an adjective and noun form in Spanish and the Spanish language has adjectives following nouns (the opposite of English), this subanalysis helped to further understand the functional way these words were used.
Another instance of non-majority was the translation for ”PTSD” in the instructions. Most kept PTSD as PTSD (36.36%) while the second most common choice spelled out PTSD as trastorno de estrés postraumático (TEPT) (24.24%). However, the goal of this translation was to translate the screen so it may also be understood by those who only speak Spanish; therefore, the majority translation of PTSD as PTSD was not used. Nor was the second choice in which PTSD was spelled out in Spanish because, in the English version of the screen, it is the disorder’s acronym that is used. Therefore, the third most frequent translation, TEPT (22.22%), was used. As a concession to aid with clarity, TEPT was used in conjunction with the full disorder name spelled out.
In the screen’s second item, the verb ”bothered” was commonly omitted (54.26%), and the next common translation referred to ”bothered” as molesto (21.28%). Omitting this word from the item significantly alters the intent of the question and arguably the construct being assessed; thus, the second most frequent translation was selected. Question seven had the phrase ”than usual” translated to the second most common translation as well. Instead of the most common translation of de lo normal (30.30%), the final choice selected was de lo usual (29.29%) in order to remain consistent with the intent of the English version’s wording. Both phrases may mean the same thing, but the common usage of ”usual” ensures a similar understanding between both language versions of this item.
There was a close tie in question nine’s translation of ”people.” Gente (45.00%) was used almost as frequently as personas (47.00%). Due to the unclear majority, a subanalysis of ”people” as it relates to the rest of the clause was conducted. ”People” was used as a part of ”people are not to be trusted.” The most popular translation of ”trusted” was [confiar]en (82.72%), and the most common translation for ”people” that used that structure for ”trusted” was personas (55.74%). This allowed the decision for the translation of ”people” to be made with greater confidence and clarity.
The rest of the decisions on which codes were used to comprise the forward translation of the screen were the wordings used by the majority of the sample.
Phase 2: Back Translation
Overall, the product from the forward translation survived the back translation with a few exceptions. The use of sentirse deprimido as ”feeling down” in the second question understandably translated literally to feeling depressed in all back translations. This is a necessary sacrifice the ITSS will have to make in the Spanish-language version. ”Feeling down” is an English colloquialism that does not mean to feel physically low. The construct being measured in this phrase is depression. Despite the participants in this phase being bilingual, they did not think to translate this phrase into a colloquialism and translated the phrase literally. To compensate for this finding, the edit was made to the ITSS-S in question two to change deprimido to abatido as it implies the intent of ”feeling down” without revealing the construct being measured (depression).
Next, for ”detached” in the fifth question of the ITSS, emocionalmente separado was translated as detached, separated, and distant. The lack of consistent translation of the word ”detached” is a weakness; however, all these translations mean the same thing semantically and would not alter how a patient would respond to this item. The use of separado is the best choice to convey the intent of this item in English. In a similar situation, ”screen” survived the back-translation process well, with the exception of the second back-translation. Within this particular translation, pantalla was translated as ”screener.” Due to the near-identical nature of ”screen” vs ”screener” and the medical appropriateness of pantalla, no changes were made to the final ITSS-S.
The last challenge was the translation for ”jumpy” in question seven. In all the back-translations, nervioso was understandably translated as ”nervous.” Similar to the issue previously discussed regarding sentirese deprimido, the majority choice from the forward translation was a literal translation of the construct being assessed by ITSS. It follows that it was a thing similarly, literally back-translated. The next top choice for ”jumpy” was asustadizo, which functions as abatido did for sentirse deprimido—it gets at the intent of the question without literally translating the construct being measured. To compensate for this finding, the translation for ”jumpy” was changed from nervioso to asustadizo.
The remaining discrepancies in the back-translated ITSS-S as compared to the original English version are minor and are the result of semantic equivalencies in wording. These differences are minor and would not lead a patient to answer the question differently as a result. Refer to Figure 1 for the resultant screen.
Fig. 1: Final ITSS-S (after back translation)
Comparison with the Medical Translation Company’s Version
The main difference between the forward translation provided by the medical translation company and the ITSS-S resulting from the study participants is different but synonymous wording (Fig. 2). Similarly stated, the company’s translation included wordings already proposed by study participants, thus indicating that the size of the study’s sample was able to ensure saturation of wordings/phrases for the screen’s headers and questions. The company’s translation did not reveal any new wordings/phrasings that were not already identified by study participants. As already described in Methods, the final translation of the screen represented the most common translations for its various elements. As such, the company’s translation did not influence deviations in wording against the greater sample of translations offered by the bilingual participants.
Fig. 2: Medical translation company’s forward translation of the ITSS
DISCUSSION
The overall goal of this study was to create a methodologically sound Spanish version of the Injured Trauma Survivor Screen (ITSS). The end-product of the Spanish translation of the ITSS (ITSS-S) survived forward and back translation with few required edits deviating from the translations offered by community and internet participants of varying degrees of bilingualism. In staying true to the language of the participants and striving to minimize, if not eliminate, investigator bias, this study achieved its aim of developing a clinically appropriate translation of the ITSS and its objective to do so in a way that maintains the original intent of the screen’s items.
Of note, the overall tone of the translation was formal. The majority of all translations utilized the formal ”you” conjugation for both the community- and internet-based samples. However, it was interesting that participants in both samples did not faithfully stick to the same pronoun’s conjugation for all ITSS items. There was not a clear pattern for when participants would switch from informal to formal, and vice versa, when conjugating for the items’ translations. A similar trend appeared when participants got into the later items of the ITSS; the formal conjugation seemed to be utilized more frequently. This trend was not explored further as it would not influence the final decision to conjugate according to the formal ”you.” This decision was supported statistically by its usage by the majority of the participants, but regardless of the outcome of that analysis, the formal ”you” would have been utilized in the end product.
Formal ”you,” or usted, is used when referring or communicating to a stranger, elders, or someone you would like to show respect to, like a physician. Informal ”you,” or tú, is used when referring or communicating to someone you know more intimately, like family and friends.18 It is most appropriate in the healthcare setting to refer to patients respectfully, making usted the natural choice for a clinical screen like the ITSS. Another reason beyond respect is linguistic clarity. There are Hispanic countries that use tú and vos equivalently, others that use only tú, and the rest that only use vos for informal second person. Usted is more universally understood, eliminates the significant potential for misunderstanding and increases the accessibility of ITSS-S to a broader, Spanish-speaking audience.
Another dialectical consideration is the wording for ”screen.” The majority translated ”screen” to pantalla. While that is the Spanish equivalent of the English word for a medical screen, it is also the Puerto Rican word for ”earring.”19,20Pantalla can also commonly refer to the screens on electronic devices or the screen that covers windows.21 Despite these common alternative meanings for pantalla, both the context of the wording provided by the rest of the screen, as well as the environment where it would be administered, should clarify the intent of the usage of pantalla as the translation of screen.
Limitations
This work is arguably limited to clinicians and researchers who interact with Latino populations that speak Mexican-derived Spanish; however, this is still an important step toward broadening access to care in the United States. Another clear limitation of this study is that this translation project is being done with a healthy community sample and the lack of opportunity to assess for construct, item, and administration bias that can only be done by administering the final product. Related to the first two points, while a necessary concession due to study design and ability to access the target population, another limitation is the lack of ability to conduct an objective assessment of the linguistic skills of the MTurk sample. Therefore, the next steps include collecting pilot data within the clinical population at Froedtert Lutheran Memorial Hospital’s Level 1 trauma center since the original ITSS screen was created for single-incident, hospitalized trauma survivors from this site. This will inform any future linguistic modifications that may be warranted regarding the wording of ITSS-S in order to give the ITSS-S the best chance of being culturally sensitive while retaining wording necessary to (1) target clinical predictors of PTSD and depression per the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, (DSM-5) within the traumatically injured population and (2) be comparable to the original English version.
Future Directions
Next steps include a pilot study to understand potential threats to validity and reliability. These threats are best summarized as internal, external, and those regarding measurement.22 Such a pilot study will also need to be conducted within the target population—traumatically injured inpatients. In addition to obtaining data regarding reliability and validity, this allows for the opportunity to understand any issues concerning the feasibility of administering the ITSS-S. The goal of the back-translation was to discover places for greater clarification, but that feedback came from preselected, healthy, highly educated, bilingual individuals and not the target population.
The result of the pilot study can then participate in a validation study to empirically test if the ITSS-S performs in a statistically comparable way to ITSS. A statistically comparable performance will enhance collaborative efforts by allowing clinicians to confidently use the ITSS in either language to predict DSM-5 criteria PTSD and depression after trauma without worrying about differences in interpretation across different institutions and across different interpreters. Additionally, comparable statistical performance would support cross-cultural research by allowing researchers to look at PTSD and depression risk for English- and Spanish-speaking subjects in aggregate rather than via separate subgroup analyses. This will eliminate a common barrier and limitation to cross-cultural studies within this area of trauma research.
If the validation study does not show comparable performance between the ITSS-S and ITSS, then this may suggest that a new PTSD and depression screen needs to be made from the ground up within the target population following the method that created ITSS (refer to Hunt et al., 2017).10 It could also suggest that the current ITSS-S needs to either remove nonpredictive items and/or add some new items that measure risk specific to this culture. Examples of new constructs to account for could be immigration and assimilation-related stress or adherence to gender roles (i.e., machismo vs marianismo).
Clinical Significance
The ITSS-S can provide immediate clinical utility for trauma centers, not only in the United States in order to comply with the ACS COT guidelines, but also in other countries/locations with Spanish-speaking patients. Prior to future pilot and validation work, this translation study provides a consistent translation of the ITSS for providers and researchers to use for their respective purposes. Having this Spanish version readily available allows for consistency by removing the possibility of differential translation by medical interpreters and research groups.
Alternative screens for PTSD after traumatic injury are limited in both English and Spanish. Common screens of PTSD include the Clinician-Administered PTSD Scale (CAPS-5),23 Primary Care PTSD Screen for DSM-5 (PC-PTSD-5),24 Posttraumatic Stress Diagnostic Scale,25 Startle, Physically upset by reminders, Anger, and Numbness Screen Derived from the Davison Trauma Scale (SPAN),26 the Short Post-traumatic Stress Disorder Rating Interview (SPRINT),27 PTSD Checklist for DSM-5 (PCL-5).28,29 Several immediate concerns are the lack of research on these tools, specifically within an injured patient sample, such as with the SPAN, SPRINT, and PC-PTSD-5. Others are not intended to predict PTSD risk. For example, the CAPS-5 semi-structured interview is intended to be a diagnostic tool administered no less than 1-month after a traumatic experience.23 The PCL-5 is a measure of symptom severity,29 though recent research has established a cutoff score at which one can reasonably conclude a patient as meeting probable diagnostic criteria.30 Also, SPAN and SPRINT were made according to the DSM-IV criteria for PTSD and have not been updated for the current DSM-5.
While there are Spanish versions available for the PCL-531 and CAPS-5,32 as well as subscales generated for specific subsets of the traumatic injury population in different countries, there is not a Spanish-specific screen currently available for PTSD risk after traumatic injury.33,34 While there does exist an automatic electronic medical record screening tool,35 even that does not replace the human element of assessing a patient’s perceived traumatic experience in their own language.36
CONCLUSION
The ITSS-S, or the Spanish-language version of the ITSS, constitutes a strong, methodical translation of a clinical risk screen for PTSD and depression after traumatic injury. As such, it has preliminary clinical utility and readiness for further validation.
ORCID
Amber Brandolino https://orcid.org/0000-0002-4232-0630
Terri A deRoon-Cassini https://orcid.org/0000-0002-9485-0625
REFERENCES
1. Johnson K. New Census Reflects Growing US Population Diversity, with Children in the Forefront. University of New Hampshire Carsey School of Public Policy. Accessed April 11, 2023. 2023. https://carsey.unh.edu/publication/new-census-reflects-growing-US-population-diversity#:~:text=The%20primary%20cause%20of%20growing,and%2030.5%20percent%20in%202000.
2. Dietrich S, Hernandez E. Nearly 68 Million People Spoke a Language Other Than English at Home in 2019. US Census Bureau. Accessed April 11, 2023. 2023. https://www.census.gov/library/stories/2022/12/languages-we-speak-in-united-states.html
3. Rotondo MF, Cribari C, Smith RS. Resources for optimal care of the injured patient. American College of Surgeons Committee on Trauma; Vol. 6. 2014.
4. American College of Surgeons. American College of Surgeons launches guidelines to help trauma centers screen patients for mental health disorders and substance misuse. Accessed April 13, 2023. 2023. https://www.facs.org/for-medical-professionals/news-publications/news-and-articles/press-releases/2023/tqp-best-practices-guideline-mental-health-substance-use/
5. Himmelstein J, Himmelstein DU, Woolhandler S, et al. Health Care Spending And Use Among Hispanic Adults With And Without Limited English Proficiency, 1999-2018. Health Aff (Millwood) 2021;40(7):1126–1134. DOI: 10.1377/hlthaff.2020.02510
6. Kundu S, Costantini TW, Doucet JJ, et al. Who is informed of trauma informed care? Patients’ primary language and comprehensiveness of initial trauma assessment. J Trauma Acute Care Surg 2023;94(1):45–52. DOI: 10.1097/TA.0000000000003815
7. Morocho C, Joplin TN, Lopez K, et al. How translators affect the experience and healthcare of primary Spanish-speaking patients in the trauma department. Proceedings IMPRS 2021;4(1). DOI: 10.18060/25712
8. Hunt JC, Chesney SA, Brasel K, et al. Six-month follow-up of the Injured Trauma Survivor Screen: clinical implications and future directions. J Trauma Acute Care Surg 2018;85(2):263–270. DOI: 10.1097/TA.0000000000001944
9. Hunt JC, Herrera-Hernandez E, Brandolino A, et al. Validation of the Injured Trauma Survivor Screen: An American Association for the Surgery of Trauma multi-institutional trial. J Trauma Acute Care Surg 2021;90(5):797–806. DOI: 10.1097/TA.0000000000003079
10. Hunt JC, Sapp M, Walker C, et al. Utility of the Injured Trauma Survivor Screen to predict PTSD and depression during hospital admission. J Trauma Acute Care Surg 2017;82(1):93–101. DOI: 10.1097/TA.0000000000001306
11. Regmi K, Naidoo J, Pilkington P. Understanding the processes of translation and transliteration in qualitative research. Int J Qual Methods 2010;9(1):16–26. DOI: 10.1177/160940691000900103
12. Saldanha G, O’Brien S. Research methodologies in translation studies. Routledge; 2014.
13. Thomas DR. A general inductive approach for analyzing qualitative evaluation data. Am J Eval 2006;27(2):237–246. DOI: 10.1177/1098214005283748
14. Saldaña J. The coding manual for qualitative researchers. 2021:1–440.
15. Amazon. About Amazon Mechanical Turk (FAQs). https://www.mturk.com/help
16. Amazon Mechanical Turk. Introducing Premium Qualifications. August 24, 2016. https://blog.mturk.com/introducing-premium-qualifications-1e473456e7b0
17. Amazon. Amazon Mechanical Turk Pricing. https://requester.mturk.com/pricing
18. Subject Pronouns in Spanish. Spanish Dictionary. https://www.spanishdict.com/guide/subject-pronouns-in-spanish
19. Earring. Tureng Dictionary and Translation, Ltd.
20. Grabb LL, McElroy OH. Merriam-Webster’s Spanish-English Medical Dictionary. Merriam-Webster; 2012.
21. Pantalla. Spanish Dictionary. http://www.spanishdict.com/translate/pantalla
22. Frey L, Botan CH, Kreps G. Investigating Communication. NY: Allyn & Bacon; 2000.
23. Weathers F, Blake D, Schnurr P, Kaloupek D, Marx B, Keane T. The clinician-administered PTSD scale for DSM-5(CAPS-5). 2013. Interview available from the National Center for PTSD at www ptsd va gov. 2016;
24. Prins A, Bovin MJ, Smolenski DJ, et al. The Primary Care PTSD Screen for DSM-5 (PC-PTSD-5): Development and Evaluation Within a Veteran Primary Care Sample. J Gen Intern Med 2016;31(10):1206–1211. DOI: 10.1007/s11606-016-3703-5
25. Foa EB, McLean CP, Zang Y, et al. Psychometric properties of the Posttraumatic Diagnostic Scale for DSM-5 (PDS-5). Psychol Assess 2016;28(10):1166–1171. DOI: 10.1037/pas0000258
26. Davidson J. SPAN Addendum to DTS Manual. New York: Multi-Health System Inc.; 2002.
27. Connor KM, Davidson JR. SPRINT: a brief global assessment of post-traumatic stress disorder. Int Clin Psychopharmacol 2001;16(5):279–284. DOI: 10.1097/00004850-200109000-00005
28. National Center for PTSD. PTSD Screening Instruments. US Department of Veterans Affairs. Updated December 6, 2022. Accessed April 11, 2023. https://www.ptsd.va.gov/professional/assessment/screens/index.asp
29. Blevins CA, Weathers FW, Davis MT, et al. The posttraumatic stress disorder checklist for DSM-5 (PCL-5): development and initial psychometric evaluation. J Trauma Stress 2015;28(6):489–498. DOI: 10.1002/jts.22059
30. Geier TJ, Hunt JC, Nelson LD, et al. Detecting PTSD in a traumatically injured population: The diagnostic utility of the PTSD Checklist for DSM-5. Depress Anxiety 2019;36(2):170–178. DOI: 10.1002/da.22873
31. Resick P, Chard K, Monson C. Cognitive processing therapy for posttraumatic stress disorder (PTSD). CPT for PTSD. Accessed April 11, 2023. 2023. https://cptforptsd.com/wp-content/uploads/2018/05/PCL-5-Spanish-version.pdf
32. Rendon MJ. The cultural adaptation of the clinician-administered PTSD Scale for Spanish-speaking Latinos with limited English proficiency in the United States. University of Miami; 2015.
33. Martínez-Levy GA, Bermúdez-Gómez J, Merlín-García I, et al. After a disaster: validation of PTSD checklist for DSM-5 and the four- and eight-item abbreviated versions in mental health service users. Psychiatry Res 2021;305:114197. DOI: 10.1016/j.psychres.2021.114197
34. Hernández-Martínez A, Martínez-Vázquez S, Rodríguez-Almagro J, et al. Validation of perinatal post-traumatic stress disorder questionnaire for Spanish women during the postpartum period. Sci Rep 2021;11(1):5567. DOI: 10.1038/s41598-021-85144-2
35. Russo J, Katon W, Zatzick D. The development of a population-based automated screening procedure for PTSD in acutely injured hospitalized trauma survivors. Gen Hosp Psychiatry 2013;35(5):485–491. DOI: 10.1016/j.genhosppsych.2013.04.016
36. Brasel KJ, Deroon-Cassini T, Bradley CT. Injury severity and quality of life: whose perspective is important? J Trauma 2010;68(2):263–268. DOI: 10.1097/TA.0b013e3181caa58f
________________________
© The Author(s). 2023 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by-nc/4.0/), which permits unrestricted use, distribution, and non-commercial reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.