Some thoughts on potential Linguistic Validation issues with US-Spanish (Part I of III)
By Henry Cifuentes
Allegedly, since we the homo sapiens species are biologically prone to distract, and err, and based on other health/medical related rationale, added to the fact that Clinical Outcome Assessment instrument translations need to be as blemish-free as possible, it is imperative to keep the human reliability (or lack thereof) factor in check.
To this end, these translations are run through a meticulous predetermined order of filtering steps meant to polish them progressively, and at the end, produce a language-and-localization pilot tested, shiny-fresh native-sounding target document. This sieving mechanism’s name is “the Linguistic Validation process.”
This is one of three posts introducing some challenges we encounter while developing a US-Spanish version of a Clinical Outcome Assessment instrument.
We will use an imaginary example to help us understand what all this entails.
ABCD Med Research Labs has developed a PRO (Patient Reported Outcome) instrument written originally in UK-English; for this instrument, they request a linguistically validated translation into US-Spanish.
The linguistic validation process starts with a Forward Translation (FT) step, where the term “forward” means moving towards the finished product (thus “forward”). For this step, we are to start by coordinating a team of professional subject-specialized translators who are native in the target language (Spanish for the US). Each of the team members submits their individual translation, and using all these versions, the team reconciles a final FT version. The next step is the BT – and yes, I think you guessed it; “BT” stands for “Backward Translation”. Why – you may ask – do we want to go “backward” when everyone nowadays prefers going “forward”?
Because by translating the instrument BACK into English-UK, and then, comparing the original source document (in UK-English) vs. its BT version (also in UK-English), we will be very surprised to find out that a few bugs, such as concept variations, missing words, among other, were able to sneak pass the FT screening step. The FT>BT cycle is concluded when the Linguistic Validation Project Coordinator (title may vary) or assignee resolves the above –mentioned issues, usually via further discussion with a Language Services Provider/Consultant who was contracted in the beginning to coordinate and supervise the process at the local level.
It is noteworthy to mention that all communication, comments, changes are logged in written format in order to maintain a traceable record of all activity. Additionally, in most cases, a “Concept List” of sorts is developed before or early in the project, and it is used to define or explain all or at least the most weight-bearing concepts, at times even providing handy alternative source wording options.
Long live the process
From here on and until the last step, there are some other optional filtering steps such as:
- Cognitive Debriefing (CogDeb or CD for short): a field pilot-test of the target document where subjects, often a panel with a good mixture of patients diagnosed with the condition in question, native speakers of US-Spanish, ideally a swath of the Spanish speaking population in the US, provide feedback strictly on their understanding of the text;
- Usability testing: if the instrument is developed for electronic use, for example an E-diary on a tablet or phone, over the Internet, etc., it is possible that a panel of subjects (generally, the same CogDeb subjects, when applicable) is appointed to field test the electronic diary, issue feedback on and rate how user‑friendly the device is. In most cases, the interviewer, in turn, observes these subjects while they try the device and reports any on questions asked, and on any minor or major roadblock they might have ran into while attempting to operate and navigate the device.
- Clinician Review: review by a subject-expert clinician (or a panel of clinicians) who is (or are) bilingual and native in the target Language (US-Spanish).
- Author’s Review: the author reviews the comments and notes in the activity logs, and makes comments, suggestions or asks questions regarding the translation or/and its process.
- Harmonization: a professional, usually one of the project coordinators/managers or assignee checks and resolves any term-harmonization issues.
- Final Proofreading: the same professional or assignee above, and quite often a second reviewer, comb the text, point out any clinging issues and resolve them.
The last cogs of the machine
- Final Review: this step usually covers a layout proof, including checking fonts, underlined, bold text and so forth.
- Approval and delivery: after any other final checks (when needed), a final approval is issued, the product is published and delivered to the end client.
The last US Census took place in 2010; however, at the rate the world is spinning in these modern days, it may seem like old data. Luckily, after the last population count, the Census has produced some estimates at various intervals. In the Profile of General Demographic Characteristics, under (or in front of) the “Hispanic or Latino” label, for July 1, 2014, displays some demographic population percentages that show 38.6% of the total inhabitants in California as “Hispanic or Latino”, 30.5% in Arizona, 38.6% [not a typo] in Texas, 24.1% in Florida, 18.6% New York State.
The term “Hispanic or Latino” (HoL for short, from here on) was coined by the U.S. Census Bureau to indicate « a person of Cuban, Mexican, Puerto Rican, South or Central American, or other Spanish culture or origin regardless of race ». This easily leads to conclude that nobody is actually native in “US-Spanish”, as their real native language is not from the US. Each of these countries plus whatever falls under “other Spanish culture or origin” has a unique history; their overall levels of income or education vary significantly, and so does ethnic (race) composition. Furthermore, many different motives and reasons drive Hispanics or Latinos to reside in the US and become part of the mainstream US system; these may include seeking education in the US, business reasons, political problems or financial strife in the home country, and so forth.
If it were possible for you to hover over the entire United States, and at a given moment, listen to the murmur of the entire gamut formed by Spanish speakers in the US, you would hear an amalgam of accents depicting significant cultural variances that play an important role in the evolution of the Spanish language in the US.
How would you define “native” US-Spanish?
Hmm, let’s see… here are some of the types of Spanish spoken in the US: Mexican, Puerto Rican, Dominican, Central American, Colombian, Chilean, Peruvian, Argentinian, and do not forget their slang-loaded cousins such as “Dominican Spanglish”, “Puerto-Rican Spanglish”, “Cuban Spanglish”, Chicano, and others. It would be a paradox or, at the very best, hardly possible that the mother tongue of an individual born in or brought to the US at a very early age, is Spanish with enough neutrality and historical uniqueness, and spoken only by a real “US-Spanish Native”, to be able to define the “US-Spanish” umbrella-term.
Spanish in the US has a short life span. It is spoken only by one or two generations, and then it quickly vanishes into the overwhelming Anglophonic influence; in other words, its main fuel comes from the large constant inflow of new immigrants.
Which Hispanics or Latinos would need or feel more comfortable with a health document that was written in their native language?
Clearly, we have to rule out all second generation US-born, US-educated HoL, and any other similar generation thereafter, because their strongest or even only language is English; when they speak Spanish, it is usually limited to close family and elders, and most notable, chances are they possess very basic or no formal education at all in Spanish.
If one of these individuals walks into a clinic and sees a version of a PRO in Spanish and another one in English, she will prefer the latter.
It is thus safe to state that “first generation HoL” who lack total command of or have limited fluency in English will be the subjects needing or preferring the PRO in its Spanish version.
These are the newcomers to the US who were born, raised, and probably schooled in countries like Cuba, Mexico, Puerto Rico, or in the South or Central American region, and whoever else falls under the “other Spanish culture or origin regardless of race” classification.
(to be continued)
The NEXT POST will deal with some of the challenges posed by translating into US-Spanish