English to Spanish translated medical forms : A descriptive genre-based corpus study 1

Medical translation in the United States has received increasing attention in recent years. This can be observed in the passing of legislation that supports equal access to healthcare regardless of language, in the development of training programs in educational institutions, and in the emergence of private initiatives in this field (Gonzalez Darriba, 2014). In this context, this paper aims to describe a genre with a very large volume of translation within medical practices, the Patient’s Personal and Medical Information Form (PPMIF), a questionnaire that gathers the relevant information needed for a successful first encounter between a patient and a medical practitioner. This description may serve as a basis for future reference when tackling the translation of PPMIF, as well as a tool for translation training and quality assessment. The analysis employed stems from a genre approach based on English for Special Purposes and Discourse Analysis applied to Translation (Gamero, 2001; García Izquierdo, 2009). Methodologically, it uses a corpus-based approach to compile and describe a parallel corpus of 100 English-language Patient’s Personal and Medical Information Forms (PPMIF) and their translations into Spanish. Subsequently, in order to provide a description for the PPMIF as a genre, a genrebased approach is applied to outline its five main components, following Gamero’s Genre Characterization Model and Jiménez-Crespo’s (2010) work on forms.


Introduction
Medical translation in the United States has received increasing attention in recent years. This new interest is reflected in new legislation that supports equal access to healthcare regardless of language, in the appearance of training programs in educational institutions, and also in the emergence of private initiatives in this field, affecting both medical translation and interpretation (Gonzalez Darriba, 2014). First, as far as legislation is concerned, according to the National Council on Interpreting in Health Care, a number of new federal laws and policies i mandate and govern the provision of interpreting services in health care settings in order to promote and ensure equal access to health care for limited English proficiency patients.
Secondly, more and more colleges, universities, professional associations, and private institutions are now offering medical translation/interpreting courses, along with certificates and training programs in this field. Lastly, a good example of a private project working towards the improvement of medical translation status is Hablamos Juntos ii , a project nationally led and funded by The Robert Wood Johnson Foundation, and whose mission is "to improve communication between health care providers and their patients with limited English proficiency." Hablamos Juntos has selected ten existing health care organizations to work with them as "demonstration sites." These demonstration sites are located in areas with high or rapidly increasing Latino population and act as experimental sites to develop and implement translation models and approaches that improve health care access for Latino patients.
The focus of these aforementioned programs and projects shows a more practical approach to medical translation in the United States; however, despite vast amounts of research in other geographical areas such as Europe or Canada, there is a scarcity of theoretical and empirical efforts to (1) describe the current situation in the US, and (2) to apply existing research in Translation Studies to its sociocultural context. This is the motivation of this paper, as it aims to categorize, describe, and analyse a genre with a very large volume of translation within medical offices: the Patient's Personal and Medical Information Form (PPMIF). Moreover, we hope this corpus and genre description will be used to inform translation quality analyses as well as translator training practices, and that they may serve as reference tools for future PPMIF translations.
Medical textual genres have been the object of research by several authors and a large number of medical genres has been identified, classified, and studied by some European research groups dedicated to this endeavour, such as GENTT iii (Géneros Textuales para la Traducción ['Textual Genres for Translation']) at Universitat Jaume I. However, genres are culture-dependent (Bazerman, 1994;Swales, 1990) and in different sociocultural contexts show different conventions or may be nonexistent all together, such as the case of the PPMIF. Therefore, careful analysis is necessary to define this textual genre in detail, as well as its conventions within the sociocultural context where it is produced.
In this paper, a description for the PPMIF is provided, drawing from previous work by Gamero (1998Gamero ( , 2001 and Devitt, Bawarshi, and Reiff (2003), and its main structural and lexical conventions are outlined. A parallel corpus of original and translated online PPMIF forms was compiled, and a descriptive characterization of the PPMIF was obtained employing Hoffmann's (1998) recurrence thresholds. The remainder of the paper is organized as follows: Section 2 introduces the theoretical framework; Section 3 presents the methodology employed to analyse the textual genre; Section 4 discusses the results, and Section 5 addresses the conclusions and limitations of the present paper.

Theoretical framework
Analysis of different types of forms (contact forms, localized web forms, medical forms, among others) has been carried out applying a variety of approaches. These include Contrastive Linguistics, Discourse Analysis, and English for Specific Purposes or Translation Studies (Grosse & Mentrup, 1980;Gülich, 1981;Jiménez-Crespo, 2010;Nord, 2001;Sánchez Nieto, 2006). These studies focused on the contrastive differences between the same type of form in two different languages (Jiménez-Crespo, 2010). As will be explained later on, this type of contrastive analysis is not possible for the medical forms under study. Instead, the use of a genre-based approach has proved useful for several reasons (Gonzalez Darriba, 2014). First of all, within the United States, medical forms share a somewhat common structure that overlaps throughout types, clinical specialties, and geographical areas -that is, medical forms present recurrent sections, moves, steps, and substeps (Swales, 1990) with similar questions driving the information exchange. Also, they are designed and organized to accomplish a certain task that must be solved within the confines of a specific communicative situation and they need to adjust to the expectations of the receiver. Finally, medical forms are a communicative instrument between patient and provider, becoming a written dialog between the participants involved in this communicative setting. The aforementioned three patterns bring to mind the notion of textual genre.
Furthermore, a genre-based approach has also been employed in Translation Studies as a helpful tool both in training future professional translators and in professional settings (Borja Albi, 2013;Borja, Izquierdo, & Montalt, 2009;Conde, 2014;García Izquierdo, 2011;García Izquierdo & Borja Albi, 2008;Hurtado, 2001;Martínez & Pugés, 2014;Montalt, Ezpeleta & García Izquierdo, 2008;Ornat, Dueñas, & Rodrigo, 2017;Orts, 2012Orts, , 2017Rodes & Mínguez, 2015, among other works). This type of approach allows for the establishment of a systematic methodology to effectively tackle the medical forms under examination, taking into account fundamental crosscultural differences in their formulation (whenever possible). Hurtado (2001) considers genre a central notion within Descriptive Translation Studies and claims that the "description and classification of textual genres are instrumental in the development of Descriptive Translation Studies, since they allow to better our knowledge of translation modalities and types, but they are also fundamental in translator education, due to their pedagogical applications" (Hurtado, 2001, p. 505). In the following section, textual genre is explained.

The notion of genre
The concept of textual genre has been widely applied to different disciplines, such as literary theory, film theory, discourse analysis or corpus linguistics, and it has been extensively defined by numerous scholars (see seminal works by Bakhtin, 1986;Halliday, 1985;Hatim & Mason, 1990). The models most relevant to Translation Studies will now be reviewed in order to understand their impact in shaping the current state of the discipline, both in theory and in practice.
Within the Linguistics field, genres have been examined from different angles, mainly Discourse Analysis (DA), Language for Specific Purposes (LSP), and Systemic Functional Linguistics (SFL) (Luzón, 2005). From DA, Bazerman (1994) framed textual genres from the viewpoint of the New Rhetoric. According to Bazerman, genre has been explored from three different perspectives: genre as text, genre as rhetoric, and genre as practice, being this last approach the one he focuses on the most. His claim is that genres are organized social actions, textual practices related to generic actions that are fundamentally dynamic, fluid, heterogeneous, and situated (Bazerman & Prior, 2005). In this description, social practices take paramount importance, over direct, straightforward textual analysis. Additionally, also from a DA perspective, Martin (2002, p. 269) defines textual genres as "configurations of meaning […] recurrently phased together to enact social practices", equally underscoring the importance of social context over textual analysis.
On the other hand, textual analysis is emphasized when examining genre through the lens of LSP or of SFL. Swales (1990) and Bhatia (1993) incorporate another factor -the discourse community -and anchor the notion of genre within that community, emphasizing the importance of the communicative action being performed by the genre and introducing the crucial role of the communicative purpose, which in turn will determine the textual lexical and grammatical features.
In an effort to integrate the different aspects set forth by the scholars previously mentioned, especially in the areas of DA and LSP, I resort to Gamero's definition (2001). Gamero integrates social practices, communicative purpose, and textual analysis in order to produce a comprehensive definition of this multifaceted concept from a Translation Studies perspective, encompassing the subtle differences present in the previous approaches while being applicable and specific to Translation Studies. From Gamero's standpoint, genre must be characterized using a dynamic model that involves an array of factors (conventional features, sociocultural context, communicative situation, text function, and intratextual elements) rather than following a rigid one-fits-all definition supposed to be adequate for every existing text. In this sense, Gamero understands genre classification in terms of prototype theory (Rosch, 1977) with instances of a certain genre being more or less central to the prototype shared by the discourse community. This description model, which will be later applied to describe the PPMIF, is presented in Figure 1: Figure 1. Genre description model from Gamero (2001, p. 60).
As seen in Figure 1, genres are framed within specific sociocultural contexts that dictate what conventions will be found in the textual exemplars. These conventions in turn mandate the intratextual elements needed to accomplish the text function in its communicative situation. Each of the defining factors included in Gamero's genre description model will be described in the following sections. Gamero (2001) employs a dynamic model in order to properly frame and describe textual genres drawing from DA and LSP. This dynamic model considers the following five pillars to be the basis of genre description: conventional features, sociocultural context, communicative situation, text function, and intratextual elements. Each of these factors is now introduced below.

Gamero's genre description model (2001)
The first pillar in Gamero's model is conventions. The fact that genres are conventionalized is the main feature that helps members of a given discourse community both produce and identify instances of textual genres, developing their genre literacy through exposure to genre exemplars (Spack, 1988). Gamero (2001) defines genre conventions as "aquellos rasgos textuales sancionados por los hablantes de una lengua y que éstos utilizan, a pesar de que existan otras opciones lingüísticamente correctas para la situación 2 " (2001, p. 51).
Gamero resorts to psychological and philosophical theories to explain how conventions happen. She follows Lewis's (1969) work on more broadly defined conventions, which claims that these repeated social norms appear when 1) several people take part in a situation, 2) the situation happens repeatedly, and 3) the people involved share a purpose only possible to achieve if they all conform to the norm. Gamero states that the communicative situations in which genres occur meet Lewis' criteria, given that: 1. The interaction involves a sender and a receiver (and maybe additional participants). 2. The situation repeats itself, since the genre format is used recurrently. 3. The participants exhibit a common behaviour and work towards fulfilling a certain purpose.
Genre conventions can manifest themselves in different ways, be it lexical choices in the text, grammatical norms or syntactical forms. As will be explained later in this study, the most common lexical conventions will be described.
The sociocultural context is the second pillar. It dictates the conventions that become recurrent in a given genre, and these conventions will need to fit the communicative situation so that the text function and purpose can be fulfilled. The sociocultural context defines the need for a genre and its influence shapes, institutionalizes or even causes particular genres within a culture to disappear. According to Gamero (1998), the effects of the sociocultural context are threefold: (1) it determines the conventions of the genres that belong to a specific social and cultural environment; (2) it is the driving force in the creation of new genres; and (3) it will determine the evolution of genres. For these reasons, some genres are common across different cultures and countries that share the need to resolve a recurrent communicative situation while other genres are bound to the society that produces them and unknown outside of that context.
The third pillar is the communicative situation, which refers to the discourse community characterized and framed by the place, time, communication mode, and identifying features of the individuals that take part in the communicative act. In other words, the communicative situation encompasses the communicative instance in which the genre is used, the text sender, the text receiver, and potential additional participants (Ezpeleta & Gamero, 2004).
The fourth pillar in defining a genre is text function, that is, the reaction that the sender wants to elicit from the receiver. As Gamero points out, a genre can accomplish more than one function at the same time, exhibiting a primary function and one or more secondary functions. This idea was previously stated by Werlich (1976), who used the term 'dominant contextual focus' to separate main and secondary functions.
According to the main intention, texts can be expositive, argumentative or instructive (Hatim & Mason, 2001). Firstly, expositive texts present information in a non-evaluative manner, focusing mainly on the description of spatial information, narration of temporal information or rendering of concepts. Secondly, argumentative texts evaluate concepts and/or beliefs and provide support or refutation for a hypothesis. Lastly, instructive texts focus on shaping future behaviour and try to influence the receiver's actions, whether giving them an option or not to act. In consequence, any given text can be primarily expositive with a secondary function, mostly argumentative with minor expositive or instructive functions or instructive in nature but with a subordinate function. Thus, the conjugation of dominant and secondary functions will shape the inherent nature of any given text.
Lastly, Gamero explains that conventions may affect the lexical and syntactical choices in a text, but at the same time, conventions may also determine intratextual features such as the textual structure and layout, what will be referred to here as 'superstructure'. Thus, conventions dictate to some extent the superstructure, contents, and their hierarchical relations in the text.
Among the range of intratextual features that can be analysed to characterize a textual genre (terminology, syntax, cohesion, tone...), researchers agree that superstructure is the most important since it is normally conventionalized in most genres and it provides the necessary cohesion to the document (Gamero, 2001;Gӧpferich, 1995). Additionally, it makes sense to apply a top-down approach to the textual analysis to subsequently complete a microstructure analysis. For these reasons, this paper focuses on the superstructure and conventional contents of the studied genre and it employs the terminology and method used by Swales (1990). The building blocks and sub-blocks of the PPMIF are to be evaluated in terms of recurrence to determine if they might be considered conventional or merely occasional.

Genre systems and medical genres
As indicated above, genres do not exist in a vacuum or as completely independent entities. Rather, related textual genres interact among them, overlapping one another and making it difficult to tease them apart, thus shaping "genre colonies" (Bhatia, 2002) or "genre systems" (Bazerman, 1994). Bazerman defines a "genre system" as "interrelated genres that interact with each other in specific settings" (Bazerman, 1994, p. 97). This concept introduces the idea of having a particular genre following upon another in a particular setting, as illustrated by the following example: a patent may not be issued unless there is an application; an infringement complaint cannot be filed unless there is a valid patent; an affidavit will not be sworn unless a challenge to the patent is filled (Bazerman, 1994, p. 98). In other words, the occurrence (or lack thereof) of each genre has consequences in the general state of affairs due to this interrelation.
In the particular case of the type of textual genre studied in this paper -a genre within the medical field -the possible "genre colony" would consist of the different instances of medical genres that have been identified to this day. Medical genres have previously been studied and defined, presenting them as effective tools for different purposes. Montalt and González Davies (2007) described several types of medical genres according to their overall social purpose (fact sheet for patients, informed consent, case report or clinical guidelines, to name a few). According to them, these texts can serve a double purpose: they may bridge the communication gap between the participants (doctor-patient, doctor-researcher, and patient-researcher) and may function as a pedagogical instrument to train new medical translators. Ezpeleta-Piorno (2012) thoroughly described the dynamic continuum of medical communication in the medicinal product information genre system. She also presented the genre shift that takes place during the intralinguistic translation of two genres within this system, that is, the translation of the summary of product characteristics into the package leaflet. In doing so, she suggests that textual competency can be acquired both by medical writers and medical translators, and crucially, these conceptualizations may aid in the development of resources and tools for the teaching and acquisition of medical writing and translation competence.
Additionally, García Izquierdo (2009) extensively researched and described the "fact sheet for patients" medical genre, and Alarcón (2016) provided an exhaustive study of health leaflets. Forés (2004) justifies the application of textual genres on translation pedagogy, and Mayor Serrano (2005) compares English and Spanish medical brochures and analyses pedagogical implications on the training of medical translators. On the other hand, Devitt et al. (2003) claim that the analysis of medical genres can give access to the discourse community that produces them, which can lead to teaching students different language uses and ways of participation in various language sites, while Salvador (2016) explores clinical case reports to describe how a practice community may determine patients' identity through their narrative. Lastly, the GENTT research group proposed a comprehensive medical genre taxonomy iv in which medical texts are classified into six categories: (1) clinical, (2) informative, (3) meta-genre, (4) educational, (5) promotional, and (6) research (Gonzalez Darriba, 2014).
Two textual genres within the medical field are explored in this paper and considered to merge into one, as it will be explained later on. Before the reasoning for this genre merge is presented, important considerations on corpus compilation are in order.

Considerations on corpora and corpus compilation
In order to understand and validate the corpus compiled for the study, basic notions in corpus studies and corpus compilation are reviewed next.
The use of corpora in Translation Studies comes from Corpus Linguistics, which at the same time stems from the definition put forth by Sinclair (1991): "a corpus is a collection of naturally-occurring language text, chosen to characterize a state or variety of a language" (Sinclair, 1991, p. 171). This definition underscores the need for collection of natural instances of language. Nonetheless, natural occurrence is not the only requirement that must be met or the only criterion to be observed when compiling a corpus. In this sense Biber, Conrad, and Reppen (1998) claim that corpora need to be both "a large and principled collection of natural texts".
Probably the most extended definition of corpus as a research entity is the one proposed by the EAGLES's (Expert Advisory Group on Language Engineering Standards) report, by which a corpus is "a collection of pieces of language that are selected and ordered according to explicit linguistic criteria in order to be used as a sample of the language" (Sinclair, 1996, p. 4). More recently, taking into consideration technological developments, Laviosa put forth that a corpus can be defined as "a collection of authentic texts held in electronic form and assembled according to specific design criteria" (Laviosa, 2002, p. 80).
In order to compile a valid and replicable corpus, Pérez Hernández (2002) points out that three main criteria must be observed: representativity, standardization, and corpus typology (with the first criterion being the one that holds paramount importance). While representativity is achieved by including in the corpus "the full range of variability in a population" (Biber, 1993, p. 243), standardization refers to the necessary filters (such as content, time and place of publication, authorship, text modality...) that must be applied to the texts collected in the compilation of the corpus. According to Laviosa's corpus typology (2002), a "bilingual parallel corpus", that is, a corpus comprised of texts in a source language A (English) and their translations in a target language B (Spanish), has been used in this study. The compilation criteria employed to ensure corpus representativity and standardization will be addressed in the section 3.2.

Methodology
The methodology for this empirical study was previously laid out in Gonzalez Darriba (2014); it follows Jiménez-Crespo (2008, 2010 and has two main components: (1) a corpus-based methodology to compile, classify, and study the textual population, and (2) an application of Gamero's Genre Characterization Model to describe the textual genre under study.

Corpus compilation, selection of the textual population, and corpus analysis
Different medical forms available online were encountered while compiling the corpus for this study. The ones shown in Figure 2 (Gonzalez Darriba, 2014) were the most commonly found forms. This intuitive classification presented in Figure 2 does not match the one put forth by GENTT due to two reasons: (1) GENTT focuses mostly on the analysis of forms in Spain and the UK, where the sociocultural context and communicative situation demand the existence of other medical genres, and (2) some of the forms encountered in the United States respond to legal requirements established by patient protection laws at the federal, state, or local level (Gonzalez Darriba, 2014). Only "New patient information" forms and "Medical history" forms were compiled and included in the corpus because they share several features that make them suitable for conveniently grouping them in a genre system: 1. In terms of content, they include very similar items, making it very hard at times to tease them apart, since most of the information asked from the patient coincides. 2. As far as the purpose v is concerned, both forms are designed to accomplish the same function: mediate and facilitate the first encounter between new patients and medical providers, so that the exchange of background information can happen successfully. 3. For many of the other forms, what is needed from the patient in order to complete the form and fulfil its purpose (conveying some policy or practice to the patient) is just a signature, "[t]here is no interaction or dialog, no questions asked and answers provided. It is as simple as getting the patients to sign their names" (Gonzalez Darriba, 2014, p. 12). However, in the case of the PPMIF, a written conversation must take place and the purpose behind the form is for the medical practitioner to gather as much information from the patient as possible. This can only be achieved with a Q&A questionnaire format where the answers to preset questions are entered by the patient, creating a written dialog that does not exist in the other form types. This particular form does not exist in cultures/countries where this dialogic interaction is realized orally.
According to these facts and based on the features they share, Gonzalez Darriba (2014) considers both the "New patient information" form and the "Medical history" form to belong to the same textual genre, hereinafter denominated "Patient's Personal and Medical Information" form (PPMIF).
For the purpose of this study, Gonzalez Darriba (2014) compiled a corpus of original English medical forms and their translations into Spanish using the World Wide Web. When compiling this bilingual corpus, the issues of representativity and standardization were addressed in the following ways: 1. Forms were included only from states with the highest percentages of Spanish-speaking populations. In order to determine which states were to be included, two variables were considered: a) percentage of population with Hispanic or Latino origins vi , and b) percentage of households where Spanish is spoken. was replaced by one of the states previously listed above. 3. The first 10 forms that belonged to the textual population under study were downloaded and included in the corpus. 4. Only forms from state-specific institutions were included in the corpus; forms from interstate and federal institutions were discarded to ensure homogeneity as much as possible. 5. Forms were only included in the corpus if the original and translated texts were separate documents, that is, forms with source and target texts presented in the same document were not taken. This decision was made because in cases where the Spanish translation of a translation unit is missing, it is impossible to discern if the translator meant to adopt the English source text as their translation or if the translator failed to translate that particular segment. However, when the original and target texts are in separate documents, the translated PPMIF can be evaluated by itself, without referencing the original English text at the translation unit level. Figure 3 offers an example of an excluded text for illustration purposes. The corpus included 100 original English forms and their Spanish translations, for a total of 200 documents. Each language contributed 94 PDF files, 4 Microsoft Word files and 2 HTML files to the corpus. The total word count is not available because some files were dead text or scanned PDFs from hard copy.
The methodology used for the analysis is based on a manual identification and count of the individual sections, moves, steps, and substeps (Swales, 1990) present in each Spanish PPMIF. Examples of these sections and moves were illustrated in Gonzalez Darriba (2014): Figure 3. English text offered first, Spanish translation added next, side by side vii (Gonzalez Darriba, 2014, p. 19) [I]t is very frequent to find forms that start with a section that asks the patient to provide some personal information. This information can then be broken down into different moves, such as basic personal information, insurance information or emergency contacts. In turn, the basic personal information move might contain steps (name, last name, date of birth...) and substeps (cellular phone number, home phone number) (Gonzalez Darriba, 2014, p. 16).
As previously applied on Jiménez-Crespo (2008, 2010 and Gamero (2001), recurrence levels are set using percentage thresholds to decide whether an item from the forms is conventional or not: • Items that appear in 70% or more of the forms: highly recurrent (Hoffmann, 1988) • Items that appear in 50% to 70% of the forms: recurrent (Hoffmann, 1988;Gamero, 1998) • Items that appear in fewer that 50% of the forms: occasional (Gamero, 1998) Applying these percentage thresholds, a prototypical superstructure and a template of the textual genre was obtained by discarding the elements that appear in fewer than 50% of the forms and keeping all recurrent and highly recurrent items. Additionally, the solutions adopted for some of the steps and substeps were explored in order to uncover the emerging trends in regards to lexical conventions in this genre. The results of these analyses are presented in Section 4.

Results: Description of the genre "Patient's Personal and Medical Information Form"
After the corpus was compiled, Gonzalez Darriba (2014) carried out a detailed analysis of the structure of all forms. The goal of this analysis was to quantitatively identify the different conventional progression of textual segments that make up this genre, that is, the sections, moves, steps, and substeps that instantiate the PPMIF. Thus, the PPMIF genre can now be described applying Gamero's genre description model.

Sociocultural context and communicative situation
The PPMIF is embedded in a sociocultural context where English is the dominant language and where Spanish speakers with English as a second language living in the US seek medical assistance for themselves or a family member at an English-speaking medical office. This causes their access to health care to become impaired due to the language barrier unless health care providers offer language-appropriate solutions. One of the options to overcome this barrier is to have translated medical forms available for this population, which can also be complemented by offering interpreting services at the office.
Furthermore, the way in which health care is organized in the United Stated requires the usage of the PPMIF while it does not exist in other countries with different health systems, where this same practitioner-patient interaction is mostly realized orally. It represents a type of intracultural mediation and therefore it is circumscribed to a particular setting. For this reason, a contrastive analysis between translated samples produced in the United States and samples spontaneously produced in Spanish-speaking countries is not possible viii .
As far as the communicative situation is concerned, Gonzalez Darriba (2014) explains that the PPMIF comes into play in the discourse community framed by the medical office the new patient enters at the time of their first visit. In this discourse community, several agents interact and engage in the information exchange (maybe more): 1. The new patient: This is the receiver of the form, who goes to the medical practice for the first time and needs to effectively communicate with the doctor in order to receive treatment. 2. The medical provider: This is the sender of the PPMIF, who needs to gather different pieces of information in order to properly treat the patient. 3. The office staff: Staff members may act as a liaison between both communication ends.
The topics and information discussed on the form are mainly medical or they may pertain to the receiver's background, and the exchange is carried out in a written manner. The language used to compose the PPMIF can be considered semi-specialized and some basic previous knowledge about the medical field is expected on the part of the patient. If the knowledge is perceived to be too specialized to be mastered by the patient, non-specialized language or some sort of explicitation is employed. For instance, an English original form from the corpus refers to seasonal allergies using the term 'hay fever', which was translated into Spanish by the translator using the Latinderived term 'polinosis'. However, the translator deems this term too difficult to be understood by the target PPMIF reader and adds an explicitation right after in brackets ("Polinosis (Alergia)").

Text function
As mentioned in section 2.1.1, texts can be classified as instructive, argumentative or expositive, depending on what their main function is. In the case of the PPMIF, Gonzalez Darriba (2014) concludes that the primary function is clearly an instructive one: "[O]nce the patient receives the form, [they] must act; the patient is expected to fill out the form as quickly and accurately as possible. By doing so, the inherently dialogic document is then complete and the text is realized" (2014, p. 23). If the patient fails to do so, the text does not accomplish the goal for which it was written in the first place. Once the form is filled, it serves a secondary expository function, and presents the patient's personal and medical information in an objective manner.
The PPMIF is designed to extract and compile the patient's personal and medical information including, and not limited to, symptoms and reasons to go to the doctor, as well as past medical history and insurance information, during the first visit to the medical practitioner. An adequate PPMIF must request all the relevant information needed for a successful appointment, as well as an appropriate diagnosis and treatment.
Although this study is not concerned with the fine print on the forms or the legal terms that are included in them, the PPMIF also serves as a legal contract and, upon the patient's signature, it becomes a binding document that links patient and physician throughout their medical and contractual interaction. By signing the PPMIF, the patient is agreeing to be treated, to share information between providers, as well as acknowledging different practices such as payment, appointment, and privacy policies. As Devitt et al. (2003, p. 551) posit, these forms are "at once a patient record, a legal document, and an element in bureaucracy, helping the doctor treat the patient and presumably protecting the doctor from potential lawsuits."

Lexical conventions
As previously mentioned in Section 2.1.1, conventions can present themselves in different ways: through lexical choices in the text, grammar or syntactic forms, text structure and layout, or content. When translating any genre exemplar, it is usually helpful for translators to have comparable target genres to use them as reference, since this allows for the replication of conventions that have been generated spontaneously in the target culture, as well as for the adequate adaptation of the source text format to one easily identified by the target users. Unfortunately, this is not the case with the PPMIF, since Spanishspeaking sociocultural contexts do not share this genre with the American sociocultural context. Consequently, the translator lacks a culturally and linguistically appropriate model to serve as a guide during the translation process and might not be able to produce an instance of a genre that embodies the set of conventional features expected or approved by the end users. This may result in a translation that is not as efficient communicatively speaking as the source text. In order to illustrate this scenario, let us employ the translation units 'Middle name' and 'Last name'.
The 'Middle name' concept is foreign to the Spanish-speaking culture and for most Spanish speakers born outside of the United States, but even so, it appears on 41% of the translated forms (Gonzalez Darriba, 2014). Although the percentage is not sufficient to make it a recurrent item, it is noteworthy how it appears on almost half of the corpus population, thus closely replicating the source text content. This lexical item may be present in the Spanish PPMIF due to a wide range of reasons. These include, but are not limited to: (1) lack of knowledge of target language cultural conventions on the part of the translator, (2) projection or generalization of source language cultural conventions on the Spanish patients (i.e., being in the United States, they may have adopted the cultural conventions, for instance, when registering a newborn baby), (3) usage of an information format consistent with computer-based data storage (that is, if a given clinic employs a software with a designated space for 'Middle name', they may prefer to have a 'Middle name' entry on their paperwork as well). These considerations make it very difficult to assess the quality or appropriateness of the observed solutions as a whole and would require a case-by-case analysis, but these are exactly the sort of disquisitions that need to happen during the translation process and in the translation classroom.
On the other hand, the majority of the translated forms that have a separate space for the last name use 'Apellido' when asking the patient's last name and only a very small percentage uses 'Apellidos' (Gonzalez Darriba, 2014). The use of 'Apellidos' acknowledges that most Spanish native speakers born in Spanish-speaking countries have two last names instead of just one (the first one being the father's first last name and the second one being the mother's first last name), as it would be expected in the American sociocultural context. Again, as with the previous lexical item, a number of reasons may be behind this decision. Some are the ones mentioned above, but there are additional factors to consider. Spanish patients may decide to drop their second last name for the sake of convenience, while others might choose to hyphenate their last names. In these scenarios, the result is the same: a single 'Apellido' is created. Moreover, as an anonymous audience member at the 2016 IMIA Annual Conference pointed out, an increasing number of Spanish-speaking individuals seem to be choosing to register their US-born children writing both last names together as a single word (i.e., Delgado Giles becomes Delgadogiles). Once more, this generates a single-word 'Apellido', so this convention may be more adequate than it was originally deemed. As with 'Middle name', it is important to keep in mind all of these potential situations when evaluating the quality of these translations, but at any rate, this is the kind of reflection that this genre characterization is hoping to inspire, in order to promote translation competence acquisition and push trainees' critical thinking skills.
As indicated by Gamero, in those instances in which more than one linguistic option is appropriate to accomplish the same communicative purpose, the choice considered to be a convention will be the one favoured by the majority of the speakers (or more specifically, the majority of the translators who translated these forms), regardless of the reason behind their choice. In the compiled corpus, these were some of the observed lexical conventions: 'Insurance'. The other two most frequent translations were 'Aseguranza' and 'Compañía de seguros'. 11. Over 60% of the forms request the patient's signature employing 'Firma del paciente'. 12. Over 80% of the forms request the date to be entered employing 'Fecha', while the rest use 'Fecha de hoy'.
Conventions not only determine lexical choices in the translated forms, but also intratextual elements, more specifically, textual structure. This feature is discussed next and a prototypical structure and template for the PPMIF will be provided shortly.

Intratextual elements: Superstructural analysis
In order to explore the macrostructure of the PPMIF, a manual identification and count of the sections, moves, steps, and substeps present in the forms in the compiled corpus was carried out. As a result of this process, the following seven major sections shown on Figure 4 were determined to be fairly common across a large percentage of the forms (arranged in order of appearance):  Although each of these major sections is comprised of a number of moves, and these in turn consist of several steps or substeps, this analysis focuses on recurrent and highly recurrent items (those that appear in 50% or more of the forms). Thus, a prototypical superstructure of the genre can be outlined. Table 1 offers the prototypical structure of the translated PPMIF, that is, the subset of textual items that meet the requirements to be considered recurrent or highly recurrent. Out of the seven major sections presented above, six reached the threshold required to be considered recurrent or highly recurrent in translated forms. Based on the prototypical structure just outlined and considering the lexical conventions previously mentioned, a PPMIF template can be created. This template, shown in Figure 5, includes all the recurrent and highly recurrent sections and moves, and exhibits the favoured lexical conventions for the steps and substeps contained in it. An interesting and worth-mentioning finding derived from the analysed corpus is that only a surprisingly small fraction of the forms includes a question regarding the patient's need for interpreting services (only 4%, to be exact). Gonzalez Darriba (2014) notes the following: It would be expected and understandable to find such question in a recurrent or highly recurrent manner since this document is geared to patients with English as a second language and designed to bridge the communication gap that exists based on language differences (p. 28).

Conclusions
This study was conducted to describe textual genre embodied by the Englishto-Spanish translated PPMIF. To do so, a corpus-based analysis was performed and Gamero's genre description model was applied to a corpus of PPMIF available online. Using this methodology, it has been established that the Spanish PPMIF exhibits a primary instructive function along with a secondary expository function. It represents a form of intracultural mediation and it is circumscribed in the US sociocultural context, where Spanish speakers with English as a second language seek medical assistance for themselves or a family member at an English-speaking medical office. It is a form that compiles all relevant information needed for a successful appointment and simultaneously acts as a patient record and a binding legal document. It comes into play within the community framed by the medical office chosen by the new patient.
The PPMIF prototypical structure and template above showed the recurrent and highly recurrent sections, moves, steps, and substeps, necessary in order to collect pertinent information for a prosperous medical practitionerpatient interaction. It also exhibits certain lexical conventions, such as 'Sexo', 'Celular', 'Seguro médico' or 'Empleador'. At the same time, these translated forms present some influence exerted by the English originals, thus incorporating lexical units that do not conform to the sociocultural norm of the target population (i.e. 'Apellido', 'Segundo nombre'). However, these observations do not intend to be a critique to the quality of the translated PPMIF samples; they rather represent a basis for reflection in order to foster the acquisition of translation competence by considering the multiple factors involved in the decision-making in the translation process and to spark productive and constructive discussions.
Limitations of the present study are related to the size of the corpus. Only one hundred forms from a limited number of states have been included in the corpus. Consequently, it is necessary to increase the size and geographical range of the collected PPMIFs in order to gain a broader understanding of this textual genre. Nevertheless, this corpus was collected with several goals in mind: (1) to be employed in later quality assessment, (2) to explore lexical variation in different geographical areas and lexical appropriateness according to the target user origin, (3) to develop educational modules to be employed in translator education, and (4) to serve as a reference tool in future PPMIF translations.
Future research is needed to address the aforementioned limitation, as well as to expand pedagogical applications, explore quality assessment, and study the cultural and theoretical implications related to the translation of forms.
viii Medical history forms might be found in other sociocultural contexts as a requirement for individuals who participate in clinical trials or research, but the form that has been analyzed in this study is specifically the one that generally accompanies "New Patient Information" forms in the first visit to a medical practice and not in other possible situations. ix Sections, moves, steps, and subteps shown in English for readers' convenience.