Writing high-quality medical publications: a user's manual

Cover
Half Title
Title Page
Copyright Page
Dedication
Table of Contents
Preface
References
	Between You and Me
	Conclusions
	References
Author
Acknowledgments: Image Credits
	Frontmatter
	Chapter 1
	Chapter 2
	Chapter 3
Chapter 1: Principles and examples of quality in medical communications
	1.1 Get Your Paper Published—On the First Attempt!
	1.2 Common Reasons Why Papers Are Rejected and Strategies to Promote Acceptance
	1.3 Unique Challenges for Medical Writers
	1.4 Principles and Examples of Ethical Medical Communications
		1.4.1 Fair balance (the foremost principle of quality) is the coin of our realm
		1.4.2 Fair balance culminates an open, free, and two-way (or “multi-way”) collaboration
		1.4.3 Fair balance in the planning process
		1.4.4 Fair balance by study report section
			1.4.4.1 Fair-Balanced Introduction
			1.4.4.2 Fair-Balanced Methods
			1.4.4.3 Fair-Balanced Results
			1.4.4.4 Fair-Balanced Discussion
			1.4.4.5 Fair-Balanced Conclusions
	1.5 Other Pillars of Quality in Medical Writing
		1.5.1 Brevity
			1.5.1.1 Lead with the News!
			1.5.1.2 Check your Gunning Fog Index
		1.5.2 Clarity
			1.5.2.1 Methods sections as the “Rosetta Stone” of manuscript clarity
		1.5.3 Cogency
			1.5.3.1 Thesis statement and topic sentences
				1.5.3.1.1 A Caveat
			1.5.3.2 Topic Sentences
			1.5.3.3 Overall Guidance on Cogency
		1.5.4 Variety
		1.5.5 Integrity
			1.5.5.1 On the other hand! Thorny factors that continue to breach integrity
				1.5.5.1.1 Lack of meaningful author participation
				1.5.5.1.2 Potential undue influence by a study sponsor
				1.5.5.1.3 The Dreaded “P Word”
				1.5.5.1.4 Other failures to disclose
		1.5.6 Credibility and fidelity
	1.6 Chapter Summary
	References
Chapter 2: Drafting the manuscript: Step-by-step guidelines and exercises
	2.1 Work Flow Dynamics
		2.1.1 Getting started
			2.1.1.2 Managing the flow of work
	2.2 Structure and Style
		2.2.1 Finding your voice: From Charles Darwin to Chris Matthews
		2.2.2 Examples of both evidence-based and memorable prose
	2.3 Structuring the Outline
		2.3.1 “Scaffolding”: The Gutkin 4 × 4 cogent manuscript structure outline
			2.3.1.1 Introduction
			2.3.1.2 Methods
				2.3.1.2.1 Internal fidelity: “Rule of Chekhov’s Gun”
			2.3.1.3 Results
			2.3.1.4 Discussion
				2.3.1.4.1 Before-after exercises in discussions (and results)
			2.3.1.5 Conclusions
	2.4 How To Write A Report of a … (“HOW–TWA–ROA”) Study
		2.4.1 Overview
		2.4.2 HOW–TWA–ROA … Systematic literature review (SLR) or meta-analysis
			2.4.2.1 Considerations When Conducting a Systematic Literature Review or Meta-Analysis and Interpreting the Findings
		2.4.3 HOW–TWA–ROA … Randomized controlled trial (RCT)
			2.4.3.1 Introduction
			2.4.3.2 Methods
			2.4.3.3 Results
			2.4.3.4 Discussion
		2.4.4 HOW–TWA–ROA … Observational study
			2.4.4.1 Overview
			2.4.4.2 Introduction
			2.4.4.3 Methods
			2.4.4.4 Results
			2.4.4.5 Discussion
		2.4.5 HOW–TWA–ROA … Health economic and outcomes research (HEOR) study
			2.4.5.1 Introduction
			2.4.5.2 Methods
	2.5 Rhetorical Exercises and “Before-After” Examples to Enhance Prose Style
		2.5.1 Motivation
			2.5.1.1 Overview
		2.5.2 Brevity
			2.5.2.1 Concise “ma non troppo”
		2.5.3 Variety
		2.5.4 General prose style
		2.5.5 Before-after examples on general prose style from my practice
	2.6 Before-After Exercises, by Manuscript Segment
		2.6.1 Article title
		2.6.2 Abstract
			2.6.2.1 ”Help Readers to Find Your Article”
		2.6.3 Acknowledgments: Before and after editing
		2.6.4 References
		2.6.5 Tables: Before and after editing
	2.7 Putting it all together: A representative (fictitious) study report for a multispecialty journal
	References
	Appendix: “Diction-Err-Y”— A guide to better usage
		“Diction-Err-Y”: Of “plural effusions” and other peculiarities of medical writing’s “singular” syntax and semantics
Chapter 3: Biostatistics: Issues in study design, analysis, and reporting
	3.1 Statistics: Sturm und Drang
		3.1.1 Micro … Starting small
	3.2 The nature and distributions of data and their implications for statistical analyses
	3.3 Signal versus noise*; Confidence versus doubt
		3.3.1 Statisticians are frequentists (unless they are Bayesian!)
		3.3.2 Overview of study designs: What we can and cannot (or should not) say in fair-balanced discussion sections
	3.4 The fundamentals: Descriptive statistics, including data distributions and measures of central tendency and dispersion (error)
		3.4.1 Data distributions and their contours
		3.4.2 Measures of central tendency
			3.4.2.1 Arithmetic mean
			3.4.2.2 Geometric mean
			3.4.2.3 Mode
			3.4.2.4 Median
		3.4.3 Measures of dispersion (“error”)
			3.4.3.1 Standard Deviation (SD)
			3.4.3.2 Variance
			3.4.3.3 Standard error of the mean (SEM)
			3.4.3.4 95% Confidence interval (CI)
			3.4.3.5 Converting 95% CI to SE and SD
			3.4.3.6 Interquartile Range (IQR)
		3.4.4 Degrees of freedom (df)
	3.5 Statistical error
		3.5.1 Types 1 and 2 Errors
			3.5.1.1 Types 1 and 2 Errors and fundamentals in computing sample size
			3.5.1.2 What p values signify (and don’t)
	3.6 Human error: Defects in logic
		3.6.1 The null hypothesis as an example of “plausible reasoning”
		3.6.2 Problems with logic: Denying the antecedent and affirming the consequent
		3.6.3 Post hoc ergo propter hoc (“after this, therefore because of this”)
	3.7 Introduction to confounding and bias
		3.7.1 Confounding
		3.7.2 Simpson’s Paradox and Lord’s Paradox
		3.7.3 Regression to the mean (RTM) and maturation effects
	3.8 Comprehensive review of precision, reproducibility, reliability, validity, effect modification, interaction, and bias
		3.8.1 Effect modification29
		3.8.2 Interaction
		3.8.3 Prominent forms of bias
			3.8.3.1 Overview
			3.8.3.2 Selection bias
			3.8.3.3 Information bias
			3.8.3.4 Other issues
	3.9 Statistical tests for categorical data: Examples and exercises
		3.9.1 Paired data
			3.9.1.1 McNemar’s test61
		3.9.2 Unpaired Data
			3.9.2.1 Fisher’s exact test (FET) or Fisher’s exact test of contingency62–64
			3.9.2.2 χ2 (Chi-squared) test66
				3.9.2.2.1 Cochran–Mantel–Haenszel (CMH)  test67,68
				3.9.2.2.2 χ2 Test of heterogeneity
	3.10 Statistical tests for continuous data: Examples and exercises
		3.10.1 Parametric tests
			3.10.1.1 Student’s (Gosset’s) t-test (smaller samples)72 or z-test (larger samples) (paired)
			3.10.1.2 One-way analysis of variance (ANOVA); F test 73,74
				3.10.1.2.1 Further background on, and implications of, ANOVA: Regression analysis, general linear model, two-factor ANOVA, repeated-measures ANOVA, and ANCOVA
			3.10.1.3 ANCOVA75
			3.10.1.4 Special examples of ANOVA
				3.10.1.4.1 Factorial ANOVA
				3.10.1.4.2­­ Multivariate ANOVA (MANOVA)
				3.10.1.4.3 Repeated-Measures ANOVA (rmANOVA)
		3.10.2 Nonparametric tests
			3.10.2.1 Wilcoxon Signed-Rank (WSR) and Rank Sum (WRS) tests80
				3.10.2.1.1 Wilcoxon signed-rank (WSR) test80
				3.10.2.1.2 Wilcoxon rank sum (WRS) test
			3.10.2.2 Friedman test82
			3.10.2.3 Mann–Whitney U test84
			3.10.2.4 Kruskal–Wallis (K–W) test85
	3.11 Selected statistical tests organized in ascending alphabetical order
		3.11.1 ANCOVA. See above (as an extension of ANOVA)
		3.11.2 Bayesian statistics87
		3.11.3 Bland–Altman (B−A) Plot93,94
		3.11.4 Bootstrapping13
		3.11.5 Correlation
		3.11.6 Correlation 1: Pearson’s product-moment correlation (coefficient = r)66,98
		3.11.7 Correlation 2: Spearman’s rank correlation (coefficient = ρ)99
		3.11.8 Correlation 3: Kendall’s correlation (coefficient = τ)101
		3.11.9 Funnel plot102
		3.11.10 General linear model (including regression models), generalized linear model, generalized linear mixed model (GLMM), and generalized estimating equation (GEE)108,109
			3.11.10.1 General linear models
		3.11.11 Heterogeneity
		3.11.12 Hosmer and Lemeshow’s test118,119 (goodness of fit in predictive models)
		3.11.13 Imputation
			3.11.13.1 Kolmogorov and Smirnov’s (K–S) and Shapiro and Wilk’s (S–W) tests96,97 and data normality
		3.11.14 Propensity score matching (PSM)
		3.11.15 Principal component analysis (PCA)
		3.11.16 Regression analyses
		3.11.17 Risk concepts: Toward a more nuanced understanding
		3.11.18 Number needed to treat to harm
		3.11.19 Survival analyses: Kaplan–Meier138 (K–M) and proportional hazards (PH)139 (or Cox regression) methods
	References
Chapter 4: Best practices: Consensus recommendations and standards to prepare high-quality, ethical, transparently disclosed manuscripts for journal publication
	Executive summary
	Part 1: QC checklists and other tables to optimize manuscript quality
	Part 2: Forms Related To Transparent And Ethical Disclosures
		Form 1
	Reference
		Form 2
		Form 3
		Form 4A
		Form 4B
	References
Appendix 1: Minimum clinically important differences (MCIDs) in patient-reported outcomes (PROs)
	REFERENCES
Appendix 2: Probability distribution data tables to compute p values
Appendix 3: Common abbreviationsa in clinical and pharmaceutical sciences
Index