Standing reference works for a medical writer

A medical writer keeps a workspace and a set of reference tools within reach of the desk, and a few standing reference works that repay reading as well as consulting: clinical and pharmacology texts, statistics and epidemiology books, grammar and usage guides, medical and general dictionaries and reference works, guides to writing up research, guides to letters and science communication, and manuals for a specialist field or tool, together with the shelves practitioners assemble from them.

Workspace and resources

A writer must not start writing before assembling what writing needs: a designated writing area, books and computer resources, and familiarity with key websites. ¹ Starting before these are in place is likely to cause frustration. ¹

The writing space needs a comfortable chair that supports the back, ideally with an adjustable height, and a well-lit surface large enough for papers, books, the computer and everything else, which may be the kitchen table. ² A computer with high-speed internet access is needed for composition, revision and submission. ² One writer finds a desktop computer with two large side-by-side screens easier for long sessions of typing, and a notebook more convenient for moving work from place to place or when a computer has to go for repair. ² Other items to keep handy are a telephone, a notepad, pens or pencils and a favourite drink. ²

Clinical and pharmacology texts

Among the language resources a medical translator, writer and editor with an almost 40-year career recommended to medical writers in 2014 are a textbook of medicine, a textbook of pharmacology and a book on statistics. ^{3, 4, 5} Harrison's Principles of Internal Medicine, then in its 18th edition, is one of the foremost textbooks on medicine in the English language, and he holds that a medical writer who does not know it should. ³ Its contributions gave him a feel for indication-specific terminology, plain medical English, collocations, the correct choice of verb for different medical and surgical procedures, avoidance of distraction, and how to compose paragraphs in medical texts. ³ Early in his career he read it for hours to familiarise himself with terminology and style, it was obligatory reading every time he faced a new indication, and it was an obligatory reference alongside medical dictionaries when looking for the exact meaning of a word or term in a particular context. ³ The search for a single word sometimes took hours, but along the way he came across other terms, definitions and turns of phrase, so the time was not wasted and added to his general knowledge of medicine. ³

Goodman and Gilman's The Pharmacological Basis of Therapeutics, then in its 12th edition, focuses on drug classes and treatments and the diseases they are used for, where Harrison's focuses on disease entities and their treatments, so the two complement each other. ⁴ He read it the same way, sometimes with a specific term in mind and sometimes to fill a quiet half hour, always learning something about language, pharmacology or medicine, and as a language reference he has learned as much from it as from Harrison's. ⁴ He calls it rewarding for any medical writer, editor or translator, and especially useful, if not indispensable, to those working in clinical pharmacology. ⁴

Neither book is cheap, but unless the latest information is needed, an earlier second-hand edition serves, and for a reader whose prime concern is language, second-hand copies are very reasonably priced. ^{3, 4} In 2014 Harrison's Online cost US$220 a year and offered information on the diagnosis and treatment of more than 4,700 diseases and disorders, continuous updating, full searchability, a drug database, videos and animations, and he calls it a staggering resource that every medical writer, editor and translator should be familiar with, whether working mainly in regulatory writing or in medical communications. ³ As far as he could see, the online Goodman and Gilman's was then available only as part of a package costing more than US$800 a year or US$50 for 48 hours, which he expected would keep most users, especially freelancers, on the paper version. ⁴ He prefers paper for both, because a few reference books on paper give valuable respite through the day from the monitor and from exclusively electronic resources. ^{3, 4} Harrison's has also been translated into many languages, and together the English and translated versions are an invaluable source of high-quality, closely reviewed translations covering the full array of medical indications. ³

The 5-Minute Clinical Consult uses an outline format to tell the busy clinician what to know about a wide variety of diseases, with an emphasis on the outpatient setting, and has a version for smartphones and tablets. ⁶ The Quick Look Drug Book, a paperback published annually, is useful for checking generic names, brand names and doses, and the Quick Look Electronic Drug Reference can be loaded on a computer as an alternative or in addition. ^{6, 7}

Statistics and epidemiology books

The book on statistics among that translator's 2014 recommendations is Tom Lang and Michelle Secic's How to Report Statistics in Medicine, then costing about €30, which gives a definition and explanation of any statistical concept a medical writer will come across, and whose subject, as its title says, is how to write about statistics. ⁵ Most medical writers have no special training in statistics and find its terminology hard to penetrate, and simple explanations of the basic concepts needed in everyday work are not easy to find. ⁵ The book can be read through to learn from or used as a reference from time to time. ⁵ Reading it teaches the language of statistics, and only when writers understand the terminology themselves can they feel confident that the reader does too. ⁵ He calls it a must for newcomers to the profession, because a non-statistician can understand what its authors are saying. ⁵ It closes with a 43-page guide to statistical terms and tests as a quick reference to the earlier chapters, which are prefaced by a detailed consideration, in simple language, of the differences between clinical and statistical significance. ⁵

A statistician working in clinical trials recommends Andrew Vickers' What is a p-value anyway? to anyone who wants to know a little more about statistics in a way that is not very boring, along with the webcomic xkcd. ⁸

Oxford University Press has published its Very Short Introduction series since 2000, and its volumes on statistics and on epidemiology may interest medical writers. ⁹ A 2014 reviewer found David Hand's Statistics largely readable and engaging, concerned with the philosophy of statistics and its central concepts and theories rather than with information of practical use, and including a defence of statisticians. ⁹ It runs from everyday applications and the consequences of data collection errors, through key terms, problems with incomplete and erroneous data, probability, and estimation and inference, to types of statistical model, with a clear explanation of overfitting, and to regression and analysis of variance. ⁹ Its laboured description of the mean seems to assume no basic knowledge of maths, while the probability chapter uses probability density without defining it and describes the Bernoulli distribution opaquely, so a reader who needed the mean defined may struggle with that chapter. ⁹

Rodolfo Saracci's Epidemiology is slightly longer and less well written. ⁹ It covers the history and development of the field, prevalence, risk, incidence and odds ratios and how to calculate them, the core study designs, systematic reviews and meta-analysis, and devotes a good few pages to determining causality, while introducing logistic regression and Cox regression only briefly, as a platform for further reading. ⁹ It assumes ignorance in its reader, writing sugar and of the lung instead of glucose and pulmonary, some of its example calculations are hard to follow, some figures are hard to interpret, several statements are hard to decode, and its descriptions of polymorphism studies and genetic variation are unconvincing. ⁹ Its insights include that monitoring patterns of Google searches for influenza can detect outbreaks faster than traditional methods, and it gives guidance on when to trust health information on the internet and in the media, highlights the ethics of data protection and of genetic analysis of biobank samples and the importance of epidemiology to clinical medicine and public health and of socioeconomic status to health at subpopulation level, and closes with a generous further-reading section pointing to textbooks, exercise books, web resources and intensive courses. ⁹ Both books seem aimed at readers with high levels of intelligence or education but little or no knowledge of the subject, and neither entirely succeeds; Epidemiology arguably provides the more useful introduction but has the more obvious weaknesses. ^{9, 10} A weakness they share is the lack of a glossary, which can be an invaluable resource for newcomers to a subject, although an excellent glossary of statistical terms can be found online. ¹⁰

A 2015 reviewer recommends reading the second edition of Richard Kay's Statistical Thinking for Non-Statisticians in Drug Regulation cover to cover, because many sections in later chapters refer back to earlier ones. ¹¹ The book is aimed at readers with a basic understanding of statistics and trial design and benefits a range of non-statistician professionals in clinical trials, including medical writers, data managers, programmers and investigators, and where necessary each chapter is accompanied by the relevant sections of the ICH E9, EMA, FDA or CHMP guidelines, which present its statistical procedures from a regulatory standpoint. ¹¹

Its first four chapters cover control groups, placebos and blinding, randomisation strategies, sampling, the normal distribution, error, and the basic statistical tests. ¹¹ Chapters five and six cover adjusting the analysis for imbalances in baseline factors, with two-way analysis of variance for continuous data, the Cochran-Mantel-Haenszel test for binary categorical and ordinal data, evaluating treatment homogeneity and the benefits of multi-centre trials, and then adjusting for several factors at once through simple, multiple and logistic regression and analysis of covariance. ¹¹ Chapter seven covers the intention-to-treat population, or full analysis set, against the per-protocol population, the dangers of compromising randomisation at the analysis stage, and approaches to missing data including complete case analysis, last observation carried forward, success-failure classification and worst-case or best-case imputation, each with the relevant ICH E9, FDA and CHMP guidance. ¹¹ Chapters eight and nine cover power and sample size in preventing type I and type II errors and how statistical significance relates to clinical significance. ¹¹

The second half covers multiplicity and multiple testing, with Bonferroni and Hochberg corrections and interim analyses; non-parametric tests where parametric ones do not apply; equivalence and non-inferiority studies, including defining confidence interval margins and using two-sided intervals for equivalence and one-sided intervals for non-inferiority; survival analysis, with censoring, Kaplan-Meier curves, event rates, median rather than mean survival and constant and non-constant hazard ratios, which the reviewer singles out for writers working mainly on oncology trials; and interim analyses, with guidance on compliance with data monitoring committees. ¹¹ The five chapters new to the second edition cover Bayesian statistics against classical and frequentist methods, including prior and posterior beliefs and the regulatory view of their use; adaptive designs, how to minimise bias in them and the regulatory guidance on their use in exploratory and confirmatory studies; non-randomised observational designs, the selection, attrition, detection and performance biases they are subject to, and the regulatory guidelines on their use; safety data analysis, the role of data monitoring committees, quantifying the benefit-risk balance for regulatory submissions and proportional reporting ratios for evaluating safety signals; and statistical methods for evaluating diagnostic methods, including receiver operating characteristic curves and the kappa statistic for agreement between two diagnostic tests. ¹² The meta-analysis chapter was restructured to add statistical methodologies, a case study and further regulatory aspects, and the book closes on the statistician's role in designing trials. ¹²

Most writers have a basic understanding of statistics, but the relevance of and rationale behind the choice of a statistical procedure may often be overlooked, and the reviewer strongly recommends the book to any medical writer who compiles clinical study reports or clinical manuscripts regularly. ¹²

The second edition of Philip Rowe's Essential Statistics for the Pharmaceutical Sciences, reviewed in 2016, addresses those who have to use statistics but have no ambition to become statisticians, and its author, who teaches in a university pharmacy department, understands that many students and researchers who need to understand statistics find it daunting. ¹³ Although aimed primarily at pharmacy students, it should be a helpful introduction or refresher for medical writers, because much medical research involves both drugs and statistics. ¹³

It uses diagrams in place of equations and relegates complex mathematical calculations firmly to the appendices, which makes the text accessible to non-mathematically minded readers and should reduce the panic that makes many of them give up and develop an allergy to statistics. ¹³ Each chapter is highly structured, opening with a concise summary of what will be covered and breaking the text up with key points in boxes and plenty of headings, so it is easy to navigate. ¹³ Black-rimmed "pirate boxes" with skull-and-crossbones icons alert readers to statistical dangers and trickery, often with humour, such as one headed "Beware of drug companies bearing Odds Ratios", and the text keeps a chatty, informal style throughout. ¹³ Rowe's premise is that many other books place far too much emphasis on the mechanical number crunching of statistical procedures, whereas most people have access to statistical software, so most problems arise from failing to use and understand statistics correctly. ¹³

The book is divided into five sections, starting with presenting data, then covering the statistical tests for continuous, nominal and ordinal data, and closing with topics such as survival analyses and questionnaires, and most of its 25 chapters cover a specific test, such as the t-test, or an aspect such as confidence intervals. ¹³ For readers who need to run statistical tests themselves, it includes guidance on entering data into common statistical software packages and a link to a free website by the author that helps with Minitab and SPSS. ¹³ It does not set out to give detailed guidance on the use of statistics or the presentation of data in medical publications, for which its reviewer's favourite resource remains Lang and Secic's How to Report Statistics in Medicine, and she suggests a future edition could add guidance on the types of graph suitable for journal articles or posters and how to produce them. ¹⁴ It offers insights into statistical language, including a section on the meaning of the term "risk", and its final chapters on multiple testing, survival analysis and questionnaires are likely to be particularly relevant to writers and include sharp criticism of the sloppy practices of many journals and the need for more rigorous statistical reporting. ¹⁴ Although the book is not written for medical writers, she thinks many would find it useful, especially those who feel they ought to know more about statistics but find other texts impenetrable. ¹⁴

Munier Hossain's Making Sense of Medical Statistics: A Bite Sized Visual Guide (Cambridge University Press, 2021; 188 pages), by an orthopaedic surgeon who teaches statistics and evidence-based healthcare to medical students, runs to 19 chapters of increasing complexity, each opening with learning objectives and closing with take-home messages, which helps a reader who wants to dip in for a quick refresher on a particular test or topic. ^{15, 16} Its final chapter uses the mishandling of coronavirus statistics in government briefings to explain how data can be manipulated during clinical trial planning and reporting, which its reviewer calls one of the best explanations of this she has seen. ¹⁶ She found its glossary almost worth having the book for alone, and judged it particularly good for medical writers new to statistics or needing a quick refresher but less useful for comprehensive coverage of statistical tests, their origins and derivations. ¹⁵

Brian S. Everitt's Medical Statistics from A to Z (third edition, Cambridge University Press, 2021; 292 pages) is aimed at those with little experience of medical research and clinical trials, and defines over 1,500 terms, 150 of them new to that edition. ¹⁷ Most definitions carry references for further reading and internal cross-references to related entries, and 88 figures accompany definitions, such as sample scatter plots and a plasma concentration time course illustrating AUC, Cmax and Tmax. ^{15, 17} Shaded boxes warn non-experts of difficulties with certain techniques, such as that single imputation of missing values invents data and so understates standard errors, where multiple imputation overcomes some of the problems. ¹⁷ A few non-statistical entries, such as "internet", look out of place and are likely left over from an earlier edition, the first having appeared in 2003, but its reviewer judged it a welcome resource for medical writers who are not statistical experts. ¹⁷

Grammar and usage guides

Two medical writers recommended in 2014 the resources they rely on for the basics of writing in English, as tried and tested, reliable and trustworthy companions that are there to help when they are needed most. ¹⁸

The first writer's five begin with A Grammar of Present-Day English, a book written for advanced students of English that she consults for day-to-day writing problems rather than the more sophisticated grammars on her shelf, and whose clear structure and many examples readers who struggle with grammatical terms will particularly appreciate; it is probably best suited to non-native speakers, and it was originally written for a German audience and gives some examples in German, but she finds it very well suited to non-Germans too. ¹⁹ The BBI Dictionary of English Word Combinations covers collocations of noun with verb, noun with adjective, adjective with adverb and verb with adverb, gives phrasal verbs, lists verbs with their prepositions and contains a number of idioms, all with examples. ²⁰ Practical English Usage is an alphabetical reference to typical problem areas for non-native writers, from brief entries on single words such as control as a false friend and presently with its different British and American meanings, to longer entries on articles, adverbs and countable and uncountable nouns. ²¹

Her last two picks were out of print in 2014 and available second hand. ²¹ The Collins Cobuild English Language Dictionary is a monolingual dictionary, the result of a language research project at the University of Birmingham, that deliberately focuses on the real usage of ordinary everyday English, with an array of well-chosen examples in every entry, special feature boxes that answer questions such as whether 12 a.m. can be written for noon, and basic grammar explained in simple language, so that it serves as a dictionary and basic grammar at once; she advises reading its introduction before using it. ²² The Synonym Finder she recommends for the sheer abundance of its synonyms as a source of inspiration for more varied and creative writing, but it suits native speakers and users with a good command of English vocabulary and style, because its synonyms mostly come without comments or explanations. ²³

The second writer's five begin with the Oxford A–Z of Grammar and Punctuation, which fits a jacket pocket and has a short, simple but comprehensive introduction to English grammar, then alphabetical keyword entries with many examples, incorrect ones included where appropriate, explanatory figures, and copious cross-references instead of repetition, all in 192 pages. ²⁴ The Oxford Guide to Plain English has a section on writing tight with many examples of striking out useless words, explodes myths such as never ending a sentence with a preposition, includes a reference list of plain English words, and defends the passive voice; its current edition had dropped the earlier general recommendation that bullet points end with a semicolon, restricting that to legal documents, and added a short section on clarity for the web. ²⁵

Bill Bryson's Troublesome Words is an alphabetised reference work that can be read straight through, pointing out a large number of common errors in modern English, from spelling to usage, without descending into nit-picking, among them who and whom, agreement of the verb with nouns such as the police, fragmentary quotations, rampant wordiness and words that are often confused; it is a complement to more basic works rather than a substitute for them. ²⁶ The Careful Writer, by Theodore M. Bernstein, the style arbiter of The New York Times and the writer of its Winners and Sinners newsletter, was issued more than 50 years before 2014 and has more than 2,000 alphabetised entries, including a balanced argument for and against different than, how to decide whether a word ending in -ics is singular or plural, and the placement of only and not in a sentence, and it has often given its recommender the confidence to defend his own opinion when it did not seem to comply with a rule. ^{27, 28} The Collins Cobuild English Guides offer comprehensive, pragmatic advice on using English for native and non-native writers alike, and a couple of them, notably the one on linking words, were out of print in 2014 but findable second hand. ²⁹

Scientific English, third edition, by Robert A. Day and Nancy Sakaduski, is a basic guide to English grammar tailored to scientists and medical and science writers, in 20 self-contained chapters addressing redundancies, abbreviations, acronyms, jargon, foreign terms, email, online publishing, blogs and writing for the web, with specific examples drawn from science rather than generic ones, a chapter on self-editing and a chapter new to that edition on electronic media. ³⁰ Its reviewer calls it an excellent book for reference. ³⁰

Medical dictionaries, abbreviations and spelling

One writer reported in 2018 that his copies of Dorland's Medical Dictionary and Stedman's Medical Dictionary were mostly collecting dust because almost everything he needed could be found online. ⁶ Stedman's Electronic Medical Dictionary, whose version 7.0 its publisher reported as defining more than 107,000 terms, is a very useful program to keep on the hard drive, although its audio pronunciations and anatomic animations add little for a writer. ³¹ With free online dictionaries available, a writer may not need a formal medical dictionary at all. ³¹

Stedman's Plus Medical/Pharmaceutical Spellchecker integrates with the Microsoft Word spell-checker and verifies medical and pharmaceutical spellings as the writer types, its latest version holds the most recently released drug names, brand names included, and words missing from it, such as Dorland's, can be added to its base. ³²

Neil Davis's Medical Abbreviations: 32,000 Conveniences at the Expense of Communications and Safety is a little-known book that helps with abbreviations a writer cannot find explained, such as TDS for traveler's diarrhea syndrome and BEGA for best estimate of gestational age, or TEMP, which can mean temperature, temporary or temporal, in ways the indexes of big reference books seldom do. ⁶

General dictionaries and reference works

A printed thesaurus need not be bought, because the one built into Microsoft Word is excellent. ³³

A reverse dictionary is the reference to reach for when the concept is clear and the word is missing, and one famous one is Bernstein's Reverse Dictionary, also by Theodore M. Bernstein. ^{28, 34} Under an ordinary word, such as drug, it lists words that describe things in it: adiaphorous, meaning neutral, for anything neither right or beneficial nor wrong or harmful, and analeptic for a drug that overcomes the effects of sedatives; under kidney sit nephritis and nephrectomy. ³⁴ It is not a thesaurus, although some of its definitions are synonyms, and it aims to stimulate the mind and give a range of ideas about words to use, such as the Socratic method for a question-and-answer method of instruction, argot or lingo for the vocabulary of a profession, and rhetorical question for a question asked for effect and not calling for an answer. ³⁴

A manuscript editor described in 2008 what he used to reach for when something in a manuscript stopped him: Bartlett's or the Oxford Dictionary of Quotations for a quotation that sounds wrong, the CRC Handbook of Chemistry and Physics for a chemical name that seems misspelled, and the Foundation Directory, a list of colleges and universities or even the Yellow Pages for an organisation with a long, unfamiliar name. ³⁵ He has five unabridged dictionaries in his office, from the 1889 one-volume Century Dictionary with its thousand-page supplement, the closest the United States has come to the Oxford English Dictionary, to the photographically reduced second edition of the Oxford English Dictionary, with several more at home and many specialised dictionaries of medicine, psychology, geology and foreign languages. ³⁵ For ordinary, work-related needs he now almost never goes to the books. ³⁵

For classical medical quotations, Maurice B. Strauss's Familiar Medical Quotations is out of print but available from online booksellers. ⁶

A standard dictionary can lack a word in use: in 2012 one writer found no entry for paraprosdokian in his Webster's Collegiate, his other dictionaries or the Oxford English Dictionary, and found it in Wikipedia, leaving open whether he had simply never met the word or it had been coined recently and not yet reached the standard dictionaries. ³⁶

Asked in 2011 for her favourite book, a veteran medical editor named the World Almanac, because it answers questions such as the currency of Bulgaria, and when told that such information is online, replied that not everyone has a computer. ^{37, 38, 39, 40} When told that not everyone has the World Almanac either, she said she was kidding and named as another favourite John Allen Paulos's A Mathematician Reads the Newspaper, a book about coincidence. ⁴¹

Guides to writing up research

How to Write and Publish a Scientific Paper, seventh edition, by Robert A. Day and Barbara Gastel, has eight parts and 41 chapters, on preliminaries, preparing the text, preparing tables and figures, publishing the paper, other writing for publications, conference communications, scientific style and other topics in communication, and an appendix on acceptable title-word abbreviations, words and expressions to avoid, and helpful websites. ^{42, 30} It is not read cover to cover, but its sections are self-contained and can be read as a topic arises, and its coverage of grant proposals and progress reports, working with the media, oral presentations and providing peer review is material not usually found in books on scientific writing. ³⁰ Its 2012 reviewer recommends it for a medical writer's library. ^{43, 30}

The first four chapters of the fourth edition of Matthews and Matthews's Successful Scientific Writing, reviewed in 2019, are on pre-writing: planning, literature searches, organising thoughts and journal selection, with a short section on writer's block that offers several suggestions and cites other publications rather than a single solution. ⁴⁴ The next four cover the general organisation of a peer-reviewed manuscript on the introduction, methods, results and discussion structure, and the parts on its periphery: title, keywords, abstract, tables and figures. ⁴⁴ About a third of the way through, the book turns to the writing itself in five chapters of about 80 pages: reporting numbers and statistics; revision, with advice such as removing multiple hedges, breaking up long sentences and removing empty fillers; style and syntax, covering tense, active and passive voice, modifiers and participles; word choice, covering jargon, nominalisation, bias-free language and a list of commonly confused words; and capitalisation and punctuation for scientific writing. ⁴⁴ The final third covers publication ethics, including plagiarism, copyright and fair use, oral presentations with visual aids, slideshows and posters, and publication itself. ⁴⁴ Since the 2008 edition the authors have doubled the number of chapters, adding publication ethics and presentations among other subjects. ⁴⁴

Beginners will find its first third on arranging a manuscript useful, while both beginners and intermediate writers may return to its sections on writing, revision, grammar and syntax. ⁴⁴ Its reviewer credits its clear style, cartoons and quotations with making it an easy read, and it gathers 30 writing exercises with an answer key in an appendix. ^{44, 45} Its high-quality spiral binding lets it lie open on a desk or a laboratory bench, as a book meant to be written in, highlighted and dog-eared. ⁴⁴

A guide addressed to beginning medical writers

Shiri Diskin's The 21st Century Guide to Writing Articles in the Biomedical Sciences (World Scientific, 2018; 132 pages), by a molecular biologist with more than 15 years as a writer, editor and reviewer of journal articles, books and regulatory documents, is written for beginning medical writers, and its 2019 reviewer finds it useful to more experienced readers as an update on current conventions and online tools. ⁴⁶

Its nine chapters follow a publication from preparation to completion. ⁴⁶ The first sets up habits such as reading in the chosen field and understanding basic research ethics, and the second covers choosing a target journal and adapting the writing to its rules. ⁴⁶ The Methods and Results come next, with what to include, paragraph structure, length, where to put auxiliary information, tense and unit selection, and the interplay of visual aids and text; then the Introduction, which defines the gap in current knowledge, outlines the hypothesis and gives the background and aims; and the Discussion, which places the study in its field, presents its strengths and limitations and gives a concrete answer to the research question. ⁴⁷ The Abstract and title follow, and the last part covers presubmission tasks, submission and postsubmission work, including deciding the lead author, handling requests for revision and interpreting editors' notes. ⁴⁷ The reviewer calls it jargon-free, accurate and up to date, rich in free online resources, and useful at every level though not encyclopaedic. ⁴⁷

A guide addressed to research students

Geoff Hart's Writing for Science Journals, reviewed in 2015, has 24 chapters on the whole manuscript process: ethics, choosing a journal, the outline, using a word processor, the sections of the manuscript in detail, experimental design, numerical and statistical considerations, figures, tables, online supplemental material, format and style, and review and publication. ¹² Its author, a fellow of the Society for Technical Communication who has worked as a scientific editor for more than 25 years and estimates having edited more than 6,000 works, includes what he calls dirty secrets: inside knowledge of how papers are reviewed and assessed for publication. ¹² Its guidelines come with the caveat that journals vary considerably and that the requirements of the target journal should be researched thoroughly. ⁴⁸

It is evidently addressed to research students, and for its reviewer that gives a medical writer awareness of the context of research writing outside the medical writer's office. ^{12, 48} Information can be hard to find, with a note on abbreviations in the chapter on the first pages while acronyms are dealt with in the materials and methods chapter, and some chapters, such as the one on using a word processor, should be little needed by a medical writer; read as a whole and navigated afterwards through the index, it is approachable and well worth consulting. ⁴⁸

A guide addressed to clinicians publishing their own work

The second edition of Daniel Byrne's Publishing Your Medical Research, reviewed in 2017, gives clinicians practical advice on writing a publishable paper in 34 chapters divided into five sections: planning, observing, writing, editing and revising. ⁴⁹ Its planning section is on designing and running clinical studies, on the rationale that journal editors and reviewers look for best practice in how trials are conducted and written up, which is best addressed by not designing flaws into the research. ⁴⁹ Its reviewer, a freelance medical writer with no influence on study design, was tempted to skim those chapters but found the finer points of trial design and of avoiding a flawed design good background, and particularly liked the panels of journal reviewers' personal views on what makes a good or bad paper. ⁴⁹

As a non-statistician she found the chapters on statistical tests short, easy to follow and informative, covering which tests should be used in which circumstances and which are not appropriate, with the reasons why, and came away understanding better what investigators and study statisticians find frustrating about how trial results should or can be presented. ⁴⁹ The chapter on multivariate analysis is more detailed and explains how it can control for confounding factors in clinical trials, which is why selecting it alongside univariate analysis can enhance research papers. ⁴⁹ Byrne also points out that statistics is evolving, so investigators must ensure they use up-to-date statistical methodology. ⁴⁹

The results chapter gives guidance on presenting statistical results and on clear and concise figures and tables, and the discussion chapter lists eight questions the discussion should answer, from pointing out the novelty in the research, through the strength of the data set and the rationale for the choice of analysis, to how and why the findings might alter clinical practice. ⁴⁹ The reviewer wholeheartedly agrees with the editing section's comments on clarity and readability and its advice on internal peer review of the paper, and finds the reproduced checklist for internal review useful in editing and revising a first draft and a table of editors' and reviewers' advice on improving writing style for impact very useful. ^{49, 50}

In some of the writing chapters, the meticulous detail that was a strength early on adds length without correspondingly greater insight, particularly where the author restates information usually covered in a journal's instructions to authors, since checking compliance with those is covered in a later chapter. ^{49, 50} The reviewer questions whether the authorship advice in the title-page chapter is entirely in line with the ICMJE recommendations reproduced in the book's own appendix, and calls the absence of any mention of good publication practice guidelines from the chapter on industry publications a major flaw that should be corrected in future editions. ⁵⁰ She would have folded the references chapter into the introduction chapter as integral to the writing process, did not find the tables of preferred replacements for pejorative terms describing patients helpful when the text already gives the most important advice, and finds that differences between UK and US English are not sufficiently accounted for. ⁵⁰ She also questions having more than 250 principles in a book meant to guide readers through an extended and complex process, since readers cannot hold them all in their heads to prompt their next action, and suggests numbered subsections instead. ⁵⁰

The book is not intended for professional medical writers and editors, who in the reviewer's opinion need not read it front to back, but its planning and observing sections may interest people without a background in clinical trials, and its writing, editing and revising sections could be a gentle introduction for new medical writers and editors, so medical writing departments in pharmaceutical companies and contract research organisations, and medical communications companies, may benefit from keeping a copy. ⁵⁰

A guide addressed to academic researchers

Yellowlees Douglas and Maria B. Grant's The Biomedical Writer (Cambridge University Press, 2018), reviewed in 2019, is aimed at academic researchers rather than professional medical writers. ^{51, 52} It has seven chapters, each beginning with a list of learning points and interspersed with information boxes headed "Between the lines" and "Snares to avoid". ⁵² Its chapter on writing for the reader's brain takes an academic approach to how people read, from word recognition to recall, with tips on organising sentences and paragraphs for impact, and the reviewer draws two principles from it: papers must be readable to have impact, and writers should always ask a copy editor to look at their work before submitting. ⁵² Its high points for her include a section on making discussions more accessible, with the BMJ's suggested structure for discussions, and an explanation of association, correlation and causation. ⁵²

A learning point of its third chapter is to anticipate potential objections or rejections while writing, which the reviewer calls good advice, and the authors remind writers to use the reporting guidelines CONSORT, PRISMA, STROBE, STARD and STREGA. ⁵² They suggest classifying a study by its outcomes, as incremental, first ever, incidental or dramatic, or a paradigm shift, while accepting that outcomes often straddle more than one category, before writing an outline, which also helps in promoting the paper after publication. ⁵¹ They recommend a thorough literature search using search algorithms in Google Scholar and PubMed, and the reviewer endorses their suggestions to create a basic outline that is gradually expanded, to use the introduction to highlight the gaps in research, to cover the who, what, when, where and how of the study in the methods, and to keep commentary out of the results, which is what journal editors and reviewers look for. ⁵²

The reviewer found the career section at the end of the third chapter troubling because it could be read as encouraging writers to flout ICMJE and GPP3 recommendations, in describing senior figures who enlist someone they mentor to collaborate on a review, where collaborate reads as write, and in noting that a mentor's name may get a paper published in a good journal, which is the behaviour the pharmaceutical industry was criticised for and that prompted those guidelines. ⁵² The authors acknowledge how harmful such arrangements are, including to women's career progression, yet encourage readers to play the game rather than change it. ⁵² She also took issue with its takeaways to consider publishing negative data, to filter the results to the most significant findings, and to highlight the most important finding in the opening paragraph of the discussion. ⁵³

Its grant-writing chapter treats an application as a business plan that leaves the reviewer no unanswered questions and cautions writers to expect to fail at the first attempt. ⁵³ Its chapter on collaborative writing, which the reviewer thinks should have come earlier, gives advice on setting and enforcing deadlines, recommends face-to-face meetings whenever practicable, especially for contentious matters, and says emails should be brief, have a clear subject, state their main purpose in the opening sentence and seek to conclude a discussion rather than start one. ⁵³ Its chapter on communicating with the public has exercises that distil a paper down to its most newsworthy findings, nine criteria of newsworthiness (impact, significance, controversy, prominence, the unusual, timeliness, currency, usefulness and educational value), and reminders to check with the publisher before issuing a press release and to publicise the research rather than any product used in it. ⁵³ The reviewer's verdict is that it is not an easy read, that some medical communications agencies may appreciate its insight into how people take in what they read, and that it is not a general reference text for professional medical writers. ⁵³

A guide addressed to social-science researchers

Pat Thomson and Barbara Kamler's Writing for Peer Reviewed Journals, reviewed in 2013, is aimed at junior social science researchers, and its authors both work in education and take most of their examples from that field. ⁵⁴ Writers of scientific papers have to disregard some of its advice, such as avoiding the IMRAD structure, and need not worry about another problem it identifies, the incapacity to insert oneself on the page. ⁵⁵ For experienced writers its points of real interest may be limited to a few pages, among them passages on avoiding signing off with an empty, lazily written conclusion and on ways of improving abstracts, illustrated with real examples. ⁵⁴ It advocates writing the abstract first, deciding what the article should say before starting to write it. ⁵⁴

Guides to letters and science communication

Listen, Write, Present, by Stephanie Roberson Barnard and Deborah St. James, is designed for medical communicators and, in its reviewer's view, is one a reader will read cover to cover, running through plan, listen, write, present, meet and serve, on the thesis that good leaders must learn to reach people outside their field as well as within it, a skill many people trained in science did not learn at school. ⁵⁶ Its section on listening teaches eye contact, listening politely, interrupting politely and accepting criticism with grace, and its reviewer says all medical writers and presenters should have it. ⁵⁶

For letters of recommendation, a medical editor recommended in 2009 Harold E. Meyer's Lifetime Encyclopedia of Letters (Prentice Hall, 1983) as all anyone could need. ^{57, 58, 59} It gives examples of virtually every conceivable kind of letter and describes ways to handle each situation, and its five-page section on letters of recommendation details ways to express an applicant's favourable attributes and, wisely and tactfully, some that are not so favourable. ⁵⁹ Its author was an industrial accountant who had spent 30 years analysing and rewording the fuzzily written letters that crossed his desk, and the book, though monumental in scope, is written in an informal style. ⁵⁹ Its contents page misspells résumé with only one, final acute accent. ⁵⁹

Manuals for a specialist field or tool

Software as a Medical Device: Regulatory and Market Access Implications by Koen Cobbaert and Gert Bos, edited by Gloria Hall (Regulatory Affairs Professionals Society, 2021; 240 pages, also as a searchable eBook), is a manual whose chapters were contributed by subject matter experts. ^{60, 61} Its 15 chapters each address a specific topic, among them the definitions that separate medical device software, software as a medical device, software in a medical device, software modules, wearables and software as an accessory, with the differences between jurisdictions; software as an in vitro diagnostic device in the EU, the US and Canada; classification; clinical evaluation of software; safety and risk management and security risk management; software development; open-source and third-party components; usability engineering; artificial intelligence; the quality and reliability of health apps that physicians need before prescribing them; and digital distribution. ^{60, 61} Writers new to the field will find the first five chapters helpful, writers experienced with devices will find much in chapters 6 to 9 in particular, and chapters 10 to 15 are too technical for writers with no experience of medical devices. ⁶² Its reviewer suggests that writers who would benefit might use it as a reference manual for answering queries rather than read it from A to Z, and its acronym list, figures, tables and index help. ^{61, 62} Medical device software is a continuously evolving topic, and new guidance may be released at any moment, particularly on artificial intelligence and online distribution. ⁶¹

Shelves practitioners assemble

Beyond the books every freelance should have, such as the AMA Manual of Style, the Merck Manual, Dorland's Illustrated Medical Dictionary and the Physicians' Desk Reference, one generalist medical writer named in 2014 four lesser-known books she finds particularly useful. ⁶³ The Writer's Lawyer: Essential Legal Advice for Writers and Editors in all Media by Ronald L. Goldfarb and Gail E. Ross is especially helpful on projects that involve coauthoring books. ⁶³ The Corporate Scriptwriting Book by Donna Matrazzo is particularly helpful for pharmaceutical sales training projects that often involve video or slide-show scripts. ⁶⁴ The Nursing Drug Handbook is simpler to use and understand than the Physicians' Desk Reference and more than adequate for most non-regulatory projects. ⁶⁴ Professional Guide to Diseases is an encyclopedia of illnesses, disorders and injuries and their treatments. ⁶⁴ Another freelance medical writer recommends The Accidental Medical Writer by Brian Bass and Cynthia Kryder to new freelance medical writers, because it offers insights that took her years to learn on her own. ⁶⁴

A medical writer working with physician-scientists and clinical researchers at a children's hospital set out in 2014 the small personal library she keeps to deepen her subject knowledge and improve her medical communication skills, noting that others' requirements may differ. ^{65, 66} Acquiring the set may seem expensive, but slightly older editions of the textbooks cost a fraction of the new retail price and may suffice for basic understanding, and anyone affiliated with a university should consult the medical library staff, because the library likely provides some electronic textbooks through an institutional login. ⁶⁶

A pharmacology text such as Basic & Clinical Pharmacology is an excellent introduction to pharmacologic concepts, useful even to writers not writing for the pharmaceutical industry, because understanding what drugs do once they enter a human body helps a writer understand what scientists hope to achieve when developing new therapies. ⁶⁷ A human anatomy and physiology text is there because one of the best ways for a writer to improve at writing about scientific research is to learn the basics. ⁶⁸ A programmed medical terminology text takes much of the mystery out of medical terminology by explaining the roots of arcane words, and used with the anatomy and physiology text gives even those without a scientific background a solid grounding in basic medical concepts. ⁶⁹ The Merck Manual of Diagnosis and Therapy is published by the pharmaceutical company of that name but is not a mere advertisement for its products, presenting nearly all medical conditions, common and rare, with objective details on currently accepted therapies, which helps a writer see where treatments fall short and so how research projects may make a difference. ⁷⁰

A project management text covers how to do the myriad tasks that let the research team do what it does best, research, while the medical writer takes care of the rest. ⁷¹ A book on intercultural communication in the global workplace is there because working with research scientists and clinicians from different cultural backgrounds is now the rule rather than the exception, and although it is oriented more to business than to scientific colleagues, its principles of how individuals communicate and the cultural reasons behind it are worth investigating. ⁷² Dennis Meredith's Explaining Research covers research articles and posters, the increasing demand for blogs, video presentations and podcasts, getting research support from stakeholders and giving inspiring oral presentations. ⁷³ Margaret Cargill and Patrick O'Connor's Writing Scientific Research Articles contains many helpful examples and uses actual journal articles as case studies, and its recommender warns that medical writers who think how-to books are beneath them ought to think again, because such books are crucial to continuing professional development as a writer and editor. ⁷⁴ On Being a Scientist, a national academies guide to responsible conduct in research, covers sometimes-thorny issues such as intellectual property, data management, conflicts of interest and allocation of author credit, with case studies to help readers apply the concepts. ⁷⁵

Useful resources for someone getting started in medical writing are too many to list, much less read, and a writer's choice of style and writing guides varies with preference and with the rhetorical requirements of the task at hand. ⁷⁶ The hospital writer's short list of general references, paper and electronic, is: ⁷⁶

• the AMA Manual of Style, whose subscription website adds search tools, updates and tips to the print version
• Author Aid, which provides instructional resources for researchers in developing countries seeking to publish and lets aspiring and new medical writers connect with scientists, many of whom speak English as a second language, and volunteer as mentors
• Thomas A. Lang's How to Write, Publish and Present in the Health Sciences, by an experienced medical writer who has taught many AMWA workshops, which like Cargill and O'Connor's book emphasises clear prose and well-designed papers
• Lang and Secic's How to Report Statistics in Medicine, which gives practical guidance on preparing reports for journals and helps writers cope with interpreting statistics
• the National Institutes of Health's grant application basics pages, which detail what to do and not to do to get research projects funded by NIH organisations, because in intense competition for dwindling research funds a well-written application is a distinct advantage
• Scientific English, with overviews of basic grammar and style, self-editing and writing for electronic media, which may be especially useful for people not fluent in English
• Scientific Style and Format, the CSE manual, whose latest edition covers publishing in print and the evolving world of online content delivery, with a subscription version holding the most up-to-date tools
• the ICMJE's Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals, formerly the Uniform Requirements for Manuscripts Submitted to Biomedical Journals, whose site anyone serious about medical writing and editing should bookmark

^{77, 78, 79, 80, 81, 82, 83, 84}

In 2014 the AMWA Amazon store listed suggested books for medical writers, including resources for beginners, books by AMWA member authors and books that have received AMWA awards. ⁸⁵

Having the right resources at hand and a comfortable work environment can make it easier for editors to contribute to creating high-quality documents. ⁸⁶ At a 2016 conference session, a veteran editor and trainer organised resources for editors into six categories, from personal comfort to business resources, and focused on those useful in the final stages of editing, which ensure correctness and consistency. ⁸⁶ Her analog tools included familiarity with style manuals, the most popular being those of the American Medical Association and the Associated Press, a custom organisational style sheet of preferred spellings and terms, and participation in local, national or global organisations for editors, which can be highly educational. ⁸⁷ Her list of websites and books added Purdue's Online Writing Lab, Grammar Girl and the OneLook dictionary search; her own blog posts on whether a freelance loses money when bidding and on time-tracking software for independents; Amy Einsohn's The Copyeditor's Handbook, Laurie Lewis's What to Charge and Steven Pinker's The Sense of Style; and, on checklists, an AMWA member's presentation on using checklists for efficient editing and Atul Gawande's The Checklist Manifesto. ⁸⁸