Let's talk more about Google Translate

Emma Goldsmith, a registered nurse-turned-Spanish to English Medical Translator, recently published an interesting piece on her blog, Signs and Symptoms of Translation: a blog by a Spanish to English medical translator. In her post, Emma briefly and comically reviewed the findings of a paper published by the BMJ. Check out her post here! We at Canopy have also previously written about the shortcomings of Google Translate here.

Further discussion of Google Translate's presence in healthcare is certainly important. To address the topic, let's do a brief refresher on the scope of language barriers in healthcare in this country:

The United States is only becoming more diverse. We have a rich history of immigration and for centuries we have provided asylum to millions forced to leave their countries. The United States has and continues to be a destination for individuals from across the globe.

The 2011 census reports that 20.8% of families in the United States speak a language other than English at home. This number is drastically larger than what it was just a few decades ago. In 1980, there were 23.1 million people speaking a language other than English at home, whereas more recently in 2010, there were 59.5 million, a 158% increase (the population grew only 38% during that time). The United States has for a long time been called “a melting pot” but the composition of the country has become increasingly diverse with a richness of different languages.

The diversity of colors and textures of the threads woven into the fabric of the nation may have become greater in number, but the systems in place are slow to adapt to properly meet the needs of the individuals hailing from other countries and cultures. Unfortunately, the availability and quality of translation services in the medical field are severely lacking. There simply are not enough interpreters available for all families to benefit from in-person human translation. Translation services are available to many practitioners over the phone but this option also has numerous downsides: elderly patients seem to be confused by the process and it is undoubtedly more difficult to translate a message without natural physical cues. Just like waiting for an in-person interpreter, waiting on hold for a phone translation will likely push the physicians to result to more immediate – and predictably inferior – unofficial translation options.

The New England Journal of Medicine published a piece in 2006 entitled, “Language Barriers to Health Care in the United States.” A case referenced in the article concerned a Spanish-speaking patient who had collapsed at his girlfriend’s residence after expressing that he was “intoxicado.” At the hospital, the patient’s girlfriend and mother repeated he was “intoxicado” which the physician translated to mean “intoxicated.” The patient was then worked up for over 36 hours for a drug overdose. The mistranslation of this one word which actually means “feeling sick to one’s stomach” undoubtedly played a part in the patient’s eventual paralysis. His true diagnosis was an intracerebellar hematoma with brain stem compression and a subdural hematoma secondary to a ruptured artery. The malpractice settlement was for settled for $71 million dollars for potentially preventable quadriplegia.

It has thus become popular – in the age of smart phones – for physicians to attempt to translate via digital means when interpreters are unavailable. Likely due to the scope of Google’s usership and the global familiarity with Google products, Google Translate has become popular in the medical community for translation with patients. It is easy to see why – upon the advent of a program that is available at a few touches for the millions with iOS or Android – practitioners are using Google translate to communicate with patients. Upon first glance, Google Translate seems like a perfect solution to the lack of availability of translation services. Across the globe, people type their queries into Google, the most used search engine in the world: “google” has been an accepted verb in the Oxford English Dictionary since 2006. “I trust Google maps, Gmail, and Google Drive! I have no reason to distrust Google translate.”

Google Translate, however, is not 100% accurate, and for conversations as sensitive as the ones one might have in a pediatric neurology clinic, terminal cancer ward, the ICU, or upon entry to the emergency department, it is vital for safety and for reasons of compassion that a medical translation is correct. The British Medical Journal published a study entitled “Use of Google Translate in medical communication: evaluation of accuracy” this past November addressing the usefulness of Google Translate for common medical statements revealing the shortcomings of the program: of the total translations, only 57.7% were correct (42.3% incorrect). Translations as sensitive and dire as those dealing with cardiac arrest, organ donation, or death were mistranslated in multiple languages.

Google Translate is not the answer. Good news is that there are alternatives to Google Translate that can be just as convenient and accessible but more accurate for medically specific translation. We at Canopy Apps are trying to provide the best alternative to a human interpreter when they are not available. The Canopy Medical Translator app has over 4,000 pre-translated medical phrases, translated by medical translators. The healthcare provider is able to select content according to specialty and can create personalized lists to aid in patient interaction. Like Google Translate, this information is available with just a few taps of a finger on a mobile device; however, unlike Google Translate – which uses machine translation to translate – Canopy’s content is accurately translated by human translators with knowledge of medical jargon.

Additionally, there is a fixed toolbar at the bottom of the screen with phone option that the healthcare provider can tap when they feel the conversation can no longer continue without an interpreter. Doing so directly connects to the telephone interpreter services provided by the user’s healthcare facility.

Thus with the Canopy system, healthcare providers can call interpreters when needed, and in situations where the do not the luxury to access human interpreters, they can rely on easy to use medical content translated into multiple languages.

With more quality tools available to solve the language barrier problem, we hope that we can make leaps in improving the health outcomes of patients who live in countries where they are often unable to communicate with those around them.

Heap | Mobile and Web Analytics