Good Writing Practice

Click to increase image sizeClick to decrease image size

Wendy Kingdom and Alistair Reeves

[email protected]; [email protected]

Cultural awareness in medical writing

A different language is a different vision of life.

Federico Fellini

In today's globalized environment, using English as the lingua franca is thought to ensure comprehensibility across audiences with different linguistic backgrounds. But is it really that simple? Even if you, as an author, use English as the agreed-upon common language, your writing will be influenced by the cultural and linguistic experiences you have been brought up with. And the same is true for your audience. They will read your text against the background of their own personal cultural experience and culturally shaped ideas, which may be very different from yours. So while you may use the same words, they may mean different things to you and your audience.¹ Talking to my French teacher, I learned that a ‘liver crisis’ (‘crise de foie’) in France has little to do with the liver, but refers to a general state of malaise, often after having had a heavy meal (and too much wine) the night before. If someone from Germany complains of ‘circulatory problems’ (‘Kreislaufprobleme’), she is suffering from low blood pressure, dizziness, and general malaise. In the UK, low blood pressure is considered as a sign of good health and ‘bad circulation’ basically means cold hands and feet. Although some of these peculiarities are anecdotal, others can indeed lead to communication problems if the author is not aware of them.

In this introductory article and further, more detailed articles on the subject, I will outline relevant cultural and language issues and suggest ways to address them.

Jargon: Jargon not only heavily relies on context and common background knowledge, but is also highly culture-specific. Stay away from it when you write for a multilingual audience. In our context, ‘jargon’ often comes in the form of ‘doctor's speak’, readily understood only by someone who has worked in an English-speaking healthcare environment (also bear in mind that American and British jargon differ considerably). Examples include terms like ‘X-plant clinic’ (transplant clinic), Ox4 (‘oriented times four’, meaning oriented in regard to person, place, time, and situation), and ‘the patient coded’ (the patient suffered cardiac arrest). If you want to read more about hospital jargon in the United States, the notorious book by Samuel Sham The House of God² offers abundant examples. As a rule of thumb, see whether you can find the relevant term in a (medical) dictionary. If not, try to find a more general term that is comprehensible to an average non-English reader. Jargon can also, however, refer to ‘common speak’ like ‘pill’ for tablets or capsules (!), or a ‘strep throat’ when simply referring to a sore throat. By the way, it is recommended to always write out angina pectoris when you refer to the cardiac condition, since ‘angina’ (without the ‘pectoris’) means ‘tonsillitis’ in German and other languages.

Which brings us to ambiguous terms: What I refer to here are terms that can easily be misinterpreted either because they have a double meaning or because of interference from other languages. Here are some examples: the term ‘alternative treatment’ for ‘other treatment’ may easily be misinterpreted in languages where ‘alternative’ (medicine, therapy, and treatment) refers to non-standard interventions like herbal medication or homeopathy. So choose ‘other’ treatment if that is what you want to say. The use of ‘should’ has been discussed elsewhere in TWS.³ You may not be aware of this, but even the simple instruction ‘Always take the tablets with your dinner’ is ambiguous because depending on their cultural background and usage, people (in the UK) may have their dinner at lunchtime or in the evening. To be on the safer side, use ‘evening meal’, even if it sounds less idiomatic. And last but not least, my pet example: while the use of ‘Caucasian’ to mean ‘White’ has become so formalized in our contexts that little ambiguity remains, you should still be aware that in some countries, including Russia, ‘Caucasian’ refers to a dark-skinned person.

To broaden our experience as medical writers in the area of jargon and potentially ambiguous terms, it would certainly make sense to start collecting similar examples from the EMWA readership. Anyone who wants to contribute on this is very welcome.

Susanne Geercken

[email protected]

Use of active and passive voice

Much has been written about this, both on paper and in the Internet, and a few misconceptions and supposed ‘rules’ have had an enormous effect on how people write in our context.

First, a warning: beware of any recommendations that reduce this issue to simple statements like ‘make more use of the active voice’ or ‘avoid the passive voice’, or even ‘don't mix the active and passive voices’. Useless advice of this sort – at least for us medical writers and editors – is given in one of the most (not really understandably) revered style guides for the English language, The Elements of Style, commonly referred to as Strunk and White.¹ The advice they give is especially bad because the five examples they give of use of the passive voice are not in the passive voice. What is obvious when you sit down and read these recommendations is that they are not intended for people writing the huge range of different types of texts in the life-science field. Also, they have been indiscriminately reproduced over the past 50 years in recommendations and style guides from all sorts of other sources.

Whether you use the active or passive voice is not just a ‘high-level’ consideration and it is not possible to give blanket advice on when one or the other should be used. Different sections of a document, for example a publication or a study report, may require different approaches. If it is unimportant whether the reader needs to know who performed a certain action – and when reporting on results in a publication or investigations in a case report, this is usually the case – then the passive is usually the best choice. If you want to bring immediacy and directness to a controversial discussion in an editorial – where ‘who said what’ is important, you will probably use the active voice to achieve this.

We shall therefore be giving guidance on this, with examples from our types of text, in future issues of Medical Writing. Readers are invited to send in any typical examples of problems with the active and passive voice so that we can use these to illustrate our recommendations. Or perhaps you would like to contribute a commentary of your own.

Alistair Reeves

[email protected]

Consistency

Using different terms does not necessarily mean being inconsistent

The word ‘inconsistent’ means not staying the same throughout a text. It also means acting at variance with one's own principles or former behaviour. In medical writing, the first definition means simply that two (or more) terms used to describe the same concept are not the same, or that different styles of presentation are used for the same elements. The second definition implies that the author has been careless, unprofessional, sloppy, and unthinking. The two definitions of inconsistent are distinct and do not necessarily apply both at the same time.

Consistency is, without doubt, very important. It is important for clarity, where use of a variety of terms is often confusing, and it is important for the professional appearance of a document. Consistency is not, however, the most important point under all circumstances, particularly in documents for regulatory use. As for many other aspects of writing, common sense and not pandering to prescriptivists should prevail.

Many examples of circumstances under which it is helpful to the reader if a rule is not applied on every occasion are similar to the rule of writing out an abbreviated term in full on the first occasion of use, providing the abbreviation in brackets, and using the abbreviation thereafter.¹ There are numerous examples where this rule can and should be broken and we have described some of these previously.^1,2 In addition, if an abbreviated term is written out in full on page 5 but the abbreviation is not used again until page 42, it is reasonable to explain the abbreviation for a second time if it is important to remind the reader what the abbreviation stands for.

If a statistician has produced a table using summary data to a mathematically correct but clinically meaningless number of decimal places (e.g. diastolic blood pressure 84.23 mmHg), presenting the data in text as 84.2 mmHg is a reasonable and sensible thing to do. One can argue that 0.2 mmHg is also clinically meaningless, however, it is reasonable to present a clinical value to one decimal place more than the measured value for summary purposes. Presenting data to more than one decimal place above the measured value will not convince anyone that the drug is effective. Clearly, the number of decimal places to which the variable is described must be the same in an in-text table and in the narrative description in the text. The point is that if the data in the statistical output tables are not sensible and the source tables will not be amended because that is too much bother, presenting rounded values in the text does not mean that the author has been inconsistent.

Similarly, if the wording in a study protocol is ambiguous or unclear, editing the wording for clarity in the study report does not mean that the report is inconsistent with the protocol. It does not make sense to reproduce words that cause confusion. However, the wording of the objectives in a protocol should not be changed no matter how badly they are written – this is going too far.

Medical terms that have been coded according to the Medical Dictionary for Regulatory Activities are written in catalogue format. It is daft to write the terms in the text exactly as they appear in the source table because we believe that consistency is the only thing that matters. Anyone who is familiar with MedDRA will know that this is like writing about pies apple or jam raspberry because this is how you would find them in the index of a recipe book. If the coded term is ‘bundle branch block right’, it is common sense to refer to it in the text as right bundle branch block. A further ‘problem’ with MedDRA is that companies in Europe have often bought the British English version but have American English as their company language. This means that the spelling of their statistical output and text differs. In such a case, it is ridiculous to alter all supportive and in-text tables manually so that they also use American spelling. A general comment should, however, be made that the spelling is different in different types of table and that alphabetization is therefore different, for example esophageal cancer and oesophageal cancer in a table or list.

Drug names are coded and the decoded terms often include the salt. There are few instances when what follows the drug name is of some relevance (e.g. isosorbide mononitrate and isosorbide dinitrate). However, in most cases, the salt is irrelevant to the pharmacology of the drug. If the document is for marketing purposes, the marketing department will be very keen to preserve brand images and trademarks and so the salt may well be included, no matter how much that irritates the reader. In, a regulatory document, with obvious exceptions such as a study comparing two salts with the same active moiety, or text in a non-clinical section on physical properties, the reader will derive the same information from reading about enalapril, as they will from reading about enalapril maleate – the latter just takes up more space. If clients, co-authors, or bosses are worried about consistency, a footnote at first mention that ‘maleate’ has been dropped should suffice.

In patient information, whether the explanatory term is given before the technical term, or vice versa, might vary according to what is being written and why. If we just want to ensure that the patient understands a technical but common term, we might write, for example nausea (feeling sick). If we are explaining something they probably would not understand, it might be better to write it the other way round, for example pain in the joints (arthralgia). This is not to say that consistency does not matter in patient information, but that there are times when an alternative approach might be better than slavishly following a rule.

‘Inconsistency’ should not be the trump card that forces writers to follow ‘rules’ even when they are not helpful for the reader. Common sense or empathy with the reader might lead the medical writer to do something differently in many cases if this is justified by circumstances. We shall be exploring further examples of this in future issues.

Wendy Kingdom

[email protected]

Lost in abbreviation: an E. coli is an EHEC is an STEC …

‘Could you put a hold on the current article you are working on and produce the following one? “E Coli (sic) Outbreaks in Europe?”’

This email, received from a US client in June 2011, started my quest of unravelling those jumbles of letters associated with a bacterium that infected thousands and killed dozens in Germany. The table below summarizes the most commonly used abbreviations in write-ups about the outbreak:

Table

EHEC vs. STEC

According to the Oxford Textbook of Medicine:

There is an important epidemiological distinction between the terms ‘EHEC’ and ‘STEC’. The former refers to STEC associated with a distinctive clinical syndrome haemorrhagic colitis, most commonly due to serotype O157:H7. Yet, other STEC can produce a range of diarrhoeal illnesses that do not fit this description. Thus, all EHEC are STEC, but only some STEC are EHEC, and STEC is a more comprehensive term.²

© Anders Holmqvist, 2012

STEC vs. VTEC

Because Shiga (named after scientist Kiyoshi Shiga, thus capitalized) toxin is synonymous with verotoxin or verocytoxin, STEC is synonymous with VTEC. However, the aforementioned dictionary states STEC is ‘more correct as it is named for the gene designation for the prototype Shiga toxin from Shigella dysenteriae type 1’.²

HUSEC, HUS STEC/HUS–STEC, and STEC HUS/STEC–HUS

HUS STEC is more specific than HUSEC. In its website, the European Centre for Disease Prevention and Control (ECDC) distinguishes between HUS STEC and non-HUS STEC cases, depending on symptom manifestation. Sometimes HUS STEC is interchangeably used with STEC HUS and I am sure the linguistics experts have a lot to say about this. It all depends on whether we are writing about the symptoms (HUS) or the pathogen (STEC). But where do the hyphens come in?

So how should we call it?

ECDC refers to it as STEC or HUS STEC. The German federal agency Robert Koch Institute which detected the first case refers to it as EHEC or EHEC O104:H4,³ the numbers being the serotype. The group of scientists who published a possible treatment calls the bug STEC and the illness STEC–HUS.⁴ The US Centers for Disease Control and Prevention (CDC) calls it STEC O104:H4⁵ to distinguish it from STEC O157:H7 that caused outbreaks in the United States in the 1990s and is under close surveillance by the CDC.

From my perspective, using the serotype is the most unambiguous way of naming this bug.

And no, I do not want to start deciphering those numbers in the serotypes. I will take the geneticists' word for it.

But whatever term you use for this bacterium, do not forget the medical writer's rule of thumb: define the abbreviation at first use and be consistent throughout the document.

Sigh. I hope I have been consistent in this text.

Raquel Billiones

[email protected]

England as the home of English – time for a rethink?

With people who speak English as a second language greatly outnumbering native speakers, where is its true home? Does it even have one?

In an article that a couple of right-leaning British newspapers picked up on,^1,2 Dr Mario Saraceni, Principal Lecturer in English Language and Linguistics at the University of Portsmouth, UK, looks to cast doubt on much of the dogma relating to the English language.³

Although many observers would link the rise of English to factors such as colonisation by the British and the ubiquity of US culture, Saraceni questions whether English can really be said to have spread from England. On what grounds? That (i) its beginnings cannot be traced to a particular place or point in time and (ii) the distinctions between languages are artificial. That is to say, there are no languages, only language.

He implicitly rejects the notion that Western English-speaking countries such as the UK and the USA should act as protectors of English, quoting fellow linguistics scholar Henry Widdowson⁴: ‘How English develops in the world is no business of native speakers in England, or the United States, or anywhere else’.

But if the UK and the USA do not own English, who does? Potentially, everyone. According to Saraceni, assigning ownership of language is not the business of academics, but rather the personal choice of individual users.

Going one step further, he describes the very concept of the native speaker as ‘flawed and misleading’, repeating Sri Lankan linguist George Braine's definition of what people perceive distinguishes native and non-native speakers⁵ as ‘country of origin, names, ethnicity, skin colour, and accent’. In other words, exclusively non-linguistic factors.

In my experience, what counts most is country of origin. This was the reason given by one potential client for rejecting an application for freelance editing assignments from my wife, who is from Sweden but (or should that be and) speaks perfect English without a hint of a Swedish accent.

Saraceni bemoans the fact that the Anglo-centric view of English has prevented other World Englishes such as Indian English and Malaysian English from gaining acceptance. The perception that it is a bastardization of true English is, he believes, to blame for the negativity of Malaysian English speakers towards Malaysian English.

For Asian, African, and other forms of English to acquire widespread recognition, he argues, the deferential ties to the supposed mother tongue(s) need to be cut. The forum for this ‘de-Anglicization of English’ must, in his opinion, be the classroom, and he devotes the second part of his article to the teaching of English.

Not altogether happy that native English speakers with no teaching qualifications can walk into TESOL (teaching English to speakers of other languages) jobs, Saraceni cites Andy Kirkpatrick of Griffith University in Australia.

Writing in 2006,⁶ Prof. Kirkpatrick criticized the practice of employing unqualified monolingual native English speakers as teachers in Japan and Korea. Mocking the assumption that they speak some form of Standard English, he contested that whatever form of English a teacher does speak may, in a country where English is not the first language, be less appropriate than the local variant.

Both Dr Saraceni and Prof. Kirkpatrick identify the use of British/American cultural reference points as a barrier to the optimal targeting of English teaching to the needs of learners. As Saraceni neatly puts it, ‘The Houses of Parliament, red double-decker buses or post-boxes, or Manhattan skylines should be confined to the realms of postcards’.

What they both fail to address is the widely held misconception that all native English speakers speak English well (according to traditional definitions). When trying to sell my own language abilities, I am always at pains to point out that am I not only a native speaker, but also an able one.

But, then, perhaps the very concept of an able native speaker is a fallacy.

Stephen Gilliver

Center for Primary Health Care Research

[email protected]

Starting sentences with prepositional phrases and clauses

Authors often create problems for themselves – or just ungainly sentences – by starting sentences with prepositional phrases or clauses where this is not necessary. Sometimes this is because they have not thought enough about the best way to express their idea, but often language interference is the culprit.

By language interference here, I mean that speakers of continental north-west European languages naturally start sentences with prepositional phrases and clauses more frequently than native English speakers – and do the same in English. Starting successive sentences with prepositional phrases and clauses often sounds rather clumsy because native English writers do this less frequently. When they do, it is often to create emphasis, as in the first sentence in this paragraph. Sometimes it is done to add variety, but much less frequently than in other languages.

The simple example below illustrates this:

[1] In an official Dutch government report issued last year (14), it is stated that 50% of inmates in preventive detention have the ICD-10 diagnosis of DPD.

Nobody will misunderstand this sentence. But was there any need to start it with in? And what is the consequence of starting with in?

With [1], we have a sentence with two clauses that starts with the subordinate (or dependent) clause. Starting with a prepositional subordinate clause often forces the author to use a dummy subject for the main clause – in this case it, but there is also frequently used. So the subject of the main clause is the dummy subject it in the middle of the sentence. The construction with it forces the clause into the passive because it is reporting on a result, which often makes sentences with dummy subjects sound clumsy.

A slight change results in sentence [2] that says exactly the same:

[2] An official Dutch government report issued last year states (stated) that 50% of inmates in preventive detention have the ICD-10 diagnosis of DPD (14).

With [2], we still have two clauses, but we have an immediately identifiable subject at the beginning of the sentence (An official Dutch government report) and have avoided the dummy subject construction by simplifying the structure of the sentence.

Another way of avoiding the prepositional phrase and dummy subject is shown in example [3]:

[3] According to an official Dutch government report issued last year, 50% of inmates in preventive detention have the ICD-10 diagnosis of DPD (14).

Here, we start with a participial phrase rather than a prepositional phrase (which should also not be overused), but the result is that it is much more likely that the author will spontaneously avoid the use of a dummy subject for the main clause and use a ‘real’ subject for the reader (50% of inmates in preventive detention) combined with the active voice.

If you start a sentence with a prepositional or participial phrase or clause, a comma is needed after the phrase or clause. As [2] shows, you avoid this by starting the sentence with the subject of one of the clauses. You also avoid this by not starting with a participial phrase or clause as in [4]:

[4] 50% of inmates in preventive detention have the ICD-10 diagnosis of DPD according to an official Dutch government report issued last year (14).

If you avoid starting sentences with prepositions, you will almost always have a simpler sentence. This does not mean that you should always avoid it. But if English is not your first language, you may do this more frequently than the reader of English expects, and it may just not sound right.

Alistair Reeves

[email protected]

Dictionaries have their uses

Looking back on another year of editing scientific text, I can offer a little advice to authors in 2012 – seeking out new words to brighten up your text is commendable but without first checking in a dictionary you do so at your peril. For instance, taking a noun and making an adjective out of it can lead to a dramatic change in meaning as in the following case where the work on synthesis becomes a fake:

The basis for the measurement of functionally active A1PI was laid by the synthetic work of Bieth et al. who described the synthesis of the convenient and water-soluble chromogenic substrate.

In another example the author muddled the expression ‘object/target of affection’ with the verb ‘affect’:

Nonetheless, it remains entirely possible that quantitative differences exist between mice and humans in terms of target organ affection.

Sometimes even dictionaries can't help. Neville Goodman (see page 68) told me that he came across the following when he was idly looking at the BMJ on line:

Professor Malcolm Green thinks the cases that have come to light are ‘likely the tip of a much larger iceberg’.

He had thought it was just the point that the tip was from a much larger iceberg, rather than a much smaller one.

Here is a word I read which you will not (yet) find in a dictionary; ‘sellness’. I came across it in the following advertisement:

Katschberg ski holidays in luxury apartment

In addition, your rental apartment includes:

•	child care services;
•	sellness and spa areas;
•	cleaning services;
•	the use of the hotel restaurant and much more.

Brightening up dull days with English

For those days when you need a laugh tonic, I can recommend a website which The Guardian in its Internet picks of the week on 23 October 2011 described as like a British version of The Onion crossed with Private Eye. The site, The Poke (http://www.thepoke.co.uk/2011/12/23/English-pronunciation/), describes itself as the product of a collective of up-and-coming comedy writers, photoshop wizards, and video mixologists – and is the fastest growing humour site in the UK. It aims to ‘deliver an ultimate antidote to the daily grind’ by ‘publishing original spoof news stories, satirical mash-ups, and brilliant photoshoppery plus the funniest stuff on the web’. Medical writers might be interested to read the English Pronunciation poem by G. Nolst Trenité which claims that ‘If you can pronounce correctly every word in this poem, you will be speaking English better than 90% of the native English speakers in the world. After trying the verses, a Frenchman said he'd prefer six months of hard labour to reading six lines aloud’.

Allie Brosh's Hyperbole and a half blog is a great example of not only good but good humourous writing, added to which the blog is copiously illustrated with delightful graphics. The home page muses over depression, the type we all get sometimes for no reason. Medical writers will appreciate the ‘Alot is Better Than You at Everything’ article about the tricks Allie, a grammatically conscientious person, uses to cope with frequently met irritating grammar mistakes. For example, when ‘you’, is written ‘u’ instead of getting mad, the economy of letters can be rationalized by imagining the person writing only has one finger on each hand. Remaining calm when ‘a lot’ is written ‘alot’ seems to be particularly difficult for Allie who has created an imaginary creature that looks like a cross between a bear, a yak, and a pug. This creature, an alot, is effective in restraining a compulsive need to correct other people's grammar to the extent that it has become almost fun for Allie to encounter the ‘a lot’ in texts. I will leave you to look at the blog yourself to see how the creature reacts to caring alot, charging alot, alot more dangerous, or liking one thing alot more than another (http://hyperboleandahalf.blogspot.com/2010/04/a lot-is-better-than-you-at-everything.html).

Elise Langdon-Neuner

[email protected]

Notes

* MedDRA; I would have included the abbreviated term here even if MedDRA has been written only once because most readers of this journal will be more familiar with MedDRA than with Medical Dictionary for Regulatory Activities. Not explaining the abbreviation MedDRA would be perfectly acceptable in a text aimed solely at regulatory and pharmacovigilance professionals