Write a title and keywords that a reader will find

A research paper's title is the first thing a reader searches for and reads. In a few words it tells what was studied, either by describing the study or by declaring its finding. Whether to name the design is a judgement, and the title and keywords together carry the terms by which a searcher finds the paper.

What the title carries

The title is the label for the paper, and it must tell, more or less, what was studied. ¹ Besides being interesting, the title is the first thing people search for. ² It is the first thing readers see, probably the first thing they read. ^{2, 3} It has to give some defined information, which might have to cover the study design, the comparators and other details the writer may not be able to control. ² Most readers of a journal look first at its index, and a title that catches their attention leads them on to the abstract, the visuals or the conclusions. ⁴

Choosing its words

A few things are best left out of a title, and beyond them the choices are a matter of style rather than guidelines. ⁵ A title should avoid rhetorical questions and carry the key words that really matter, and where the data are good it should put them forward by showing what was proven or shown. ⁵ Long, wordy titles are best avoided, since they add no meaning, and a title is better kept short. ^{5, 6} The title may be limited in length, and the instructions for authors may prescribe a word or character limit. ^{6, 1} Editors generally counsel against witty titles built on clever phrases. ¹

Each word of a title counts and must be important, like a chocolate in a limited box, and the writer should think about the power of every word. ^{3, 6, 5} A title should avoid too many prepositions and joining words that add no meaning, and carry only the really important words: those that cover the study design or what matters most in the key message of the study. ⁶ Indexing words or keywords may also be included where appropriate, depending on the journal and the search strings available. ⁶ Which words go in depends on whether clinicians or researchers read the paper and on what matters more to them, such as that a drug is given as an oral dose or that it improves 12-month survival. ⁷

A title should entice readers and be as interesting as possible. ^{6, 8} It should come back to what matters most about the article, such as quality of life, survival or the drug itself, and give a taste of its key message. ⁸

Naming the study design

Some journals insist on or suggest the study design in the title. ⁷ One guide advises including the design where possible. ⁹ Where the choice is the writer's, another view is that the writer should ask whether stating the design is interesting, such as whether the title really needs to say that the study is case-controlled. ^{8, 10} It might matter where that design is the only type of research that can be done in the area, or where it is the first time it has been done, and at some stages stating it may not be appropriate. ^{8, 10, 7}

When the title is written

The title should probably be the last thing written. ³ It is likely to be written once the paper is written and the story is understood, and it should then pull together the key words that have to go in and introduce the story. ⁵ One view is to use the key message of the article as the title while writing, which helps keep the writing on track, and to judge towards the end whether to keep it. ⁷

Declarative and descriptive titles

An early question the writer must answer is whether to reveal the conclusion in the title. ¹ A title that tells the conclusion is a declarative title, as in Genetic Evidence Found for Causal Relationships Between Maternal Obesity-Related Traits and Birth Weight, which inserts a verb. ¹ One view holds that readers need a verb in the title, as a newspaper headline usually has, and that to be meaningful the title should convey the message, as in Rickettsial Endocarditis Is Not A Rare Complication Of Congenital Heart Disease In Dental Practice: A Report Of Five Cases. ¹

A reader with no background knowledge who sees only the title Association Between Prescription of Major Psychotropic Medications and Violent Reoffending After Prison Release might presume that the medications lead to violent reoffending, when the study showed the opposite, and one view holds that the declarative Prescription of Major Psychotropic Medications Reduces Violent Reoffending After Prison Release would be a better title. ¹ The same view asks whether Aircraft Noise, Air Pollution, And Mortality From Myocardial Infarction might be improved by stating that there does seem to be an association between aircraft noise and myocardial infarction. ¹

Some journals encourage declarative titles, while others discourage or even prohibit them. ^{1, 11} A study found no evidence that a declarative title affected readers' perceptions of a study's conclusions. ¹

Advice on declarative titles differs. ^{9, 11} A guide to high-quality medical publications advises against titles that are declarative sentences conveying the study's results, because they can reduce the reader's motivation to read the paper. ⁹ On its advice, the declarative Treatment with spastex improves diaphragmatic function without causing off-target anticholinergic effects in patients with involuntary noisome hiccup syndrome: a randomized controlled trial is to be avoided, a question such as Can treatment with spastex improve diaphragmatic function without causing off-target anticholinergic effects in patients with involuntary noisome hiccup syndrome? A randomized controlled trial is acceptable, and a descriptive title such as Effects of treatment with spastex on diaphragmatic function in patients with involuntary noisome hiccup syndrome: a randomized controlled trial is the simplest and preferred. ⁹ A guide to strategic scientific and medical writing holds instead that the published title should clearly reflect the nature of the drug and the outcome of the study to entice the reader, as in First human study with a novel oral renin antagonist for the treatment of hypertension supports further development, and that a study protocol's title, usually boringly descriptive, should not be used as the paper's title. ¹¹ For a journal that does not accept conclusion-type titles, the title could drop the outcome, as in First human study with a novel oral renin antagonist for the treatment of hypertension. ¹¹

Syntactic patterns of the title

The title is the most circulated section of a journal article, and its syntactic pattern ranges from the traditional descriptive type to the assertive type, through six patterns, each shown on the same topic. ^{12, 13, 14}

1. Post-noun modifiers, a noun phrase followed by one or more prepositional phrases: Intergenerational linkages for timing of family life course events. The pattern is traditional, but its consecutive prepositional phrases impede immediate comprehension, and descriptive titles are often preceded by superfluous words such as studies on, observations on or the effect on.
2. Pre-noun modifiers, a noun preceded by one or more adjectival nouns: Family life course intergenerational linkage timing. This pattern is also traditional, and it too impedes immediate comprehension, here through stacked pre-nouns rather than consecutive prepositional phrases.
3. Coordinated nouns: Family life course intergenerational linkage and timing. The inexplicit relation between the coordinated nouns is distracting.
4. A question, a semi-assertive form: How are family life course intergenerational linkages coordinated with timing? The style is thought-provoking and interesting and could be useful for a poster or a presentation at a meeting, but it is wordy, which distracts, and it is non-traditional as a title for a journal article.
5. A suspended title: Family life course intergenerational linkages: Coordinated with timing. The pattern is non-traditional: key content words on the topic are followed by the assertive message, and a colon replaces an interrelation marker such as a preposition, but the colon between title and subtitle could lack clarity.
6. A declarative sentence: Family life course intergenerational linkages can be coordinated with timing. The style is immediately comprehensible, like a mini-abstract, but it is wordy and can be overly assertive for a controversial topic.

^{12, 15, 16, 17, 13, 14, 18, 19, 20, 21, 22, 23, 24}

The possible distractions in the first five patterns and the lack of assertiveness in the first four may be the reason for the tendency in 2024 to use the declarative sentence despite its wordiness. ²⁵

Colons, acronyms and citation rates

Colons often appear in article titles. ¹ A colon allows a progression from the general to the specific within one title, as in A July Spike in Fatal Medication Errors: A Possible Effect of New Medical Residents, which names the general problem and then the research approach, so the reader has a better idea of the contents than the first phrase alone would give. ¹

It is fashionable to give a research study an acronymic name and carry it into the title of its report, as in Effect of Laparoscopic-Assisted Resection vs Open Resection on Pathological Outcomes in Rectal Cancer: The ALaCaRT Trial, whose acronym comes from the Australian Laparoscopic Cancer of the Rectum Trial. ¹ The acronym BELT in Anticholinergic vs. Long-Acting Beta-agonist in Combination With Inhaled Steroids in Black Adults with Asthma: the BELT Randomized Clinical Trial comes from The Blacks and Exacerbations on LABA vs Tritropium (BELT) Study, and one view holds that it seems to be a bacronym, a word turned into an acronym by creating a phrase to match its letters. ¹ One view holds that titles should not contain acronyms or abbreviations, however widespread the author and editor consider their use. ¹

A title can influence how often the article is cited. ¹ Perhaps surprisingly, articles with longer titles tend to be cited more than those with shorter ones. ²⁶ In a study of articles in three prestigious journals, the number of citations correlated positively with the length of the title, as it did with a colon in the title and an acronym, and a reference to a specific country in the title predicted poor citation. ¹

A title that search engines find

What matters most for an article online is that people find it, so its title should be one that search engines find. ²⁶ A machine makes the first sort and people the second, so clever puns belong in subheadings. ²⁶ The best titles are narrative ones that capture the essence of the article and include keywords. ²⁶ An informative title that contains relevant keywords is a key requisite for a reader doing a literature search to find the paper and read it. ⁴ Settling a title is worth time: the writer should trial it by running searches in various engines and be prepared to refine it. ²⁶

People search on key phrases, not just single words, such as women's health rather than health, and the title should contain the main key phrase for the topic and be clear, descriptive, unambiguous and accurate, and read well. ^{27, 28} Search engine optimisation advice from 2023 favours a short title carrying one or two keywords, with a keyword in the first 65 characters, because most search engines shorten long titles in their listings, and for the same reason it keeps catchy phrases and puns to the subtitle. ²⁹ Special characters such as &, * and / and diacritical marks are best kept out of the title, because papers whose titles contain them are often cited incorrectly. ³⁰

An article on youth anti-war protests titled Peace Children could not be found in Google Scholar by searches on a variety of phrases around its subject. ^{31, 32} Its title meant nothing outside the printed issue and lacked key phrases such as youth anti-war protests, and the article listed only two keywords, neither of them one of its key phrases. ^{33, 32} An article titled Treatment of depression: Time to consider folic acid and vitamin B 12 came top of a Google Scholar search on depression and folic acid. ^{34, 35} Its clear, descriptive title carried the main key terms, and its keyword field repeated them. ^{36, 35} Many other factors influence search ranking. ^{33, 36}

Keywords

Keywords can keep a report from being lost, and as part of its retrievability they can contribute to the number of times it is cited. ³⁷ Many journals ask in their instructions for authors for a short list of keywords or phrases, which are used to cross-index the article and may or may not be published with the abstract. ³⁷ Terms from the Medical Subject Headings (MeSH) should be used where possible. ³⁷

Keywords should have a high likelihood of matching the terms used by someone doing a literature search. ³⁸ Important keywords should include the disease, the target organ or tissue, the name and class of the drugs studied, and words for the outcome, such as survival, toxicity, cure and progression. ³⁸ A paper can contain important information on a method although its aim was not to develop or evaluate one, and it is then useful to give it keywords that let someone looking for methods find it in a literature search. ³⁸

Identifying keywords is the easiest place to start in optimising an article for search engines. ³⁹ Most authors leave the choice of keywords until just before submission, which is a mistake, because keyword selection and use is fundamental to search engine optimisation and extends beyond the three to five words journals often ask for. ³⁹ The choice starts from the target audience and the terms it would search with, which will likely differ between a healthcare practitioner and a member of the public. ³⁹ Typing candidate terms into Google then shows which related phrases are frequently searched. ³⁹ A writer can ask what key phrases they would give a search engine if they were searching for their own article. ²⁸ Keywords should describe the content of the article, not its results, and they should be singular rather than plural. ⁴⁰ Some journals still require them to be Medical Subject Headings, while others accept any relevant words or phrases. ⁴⁰ Spelling needs weighing too: more people search in US English, although the journal often decides the language. ⁴⁰ Once the keywords are chosen, the writer should try to use them in the title. ⁴⁰

The keywords should reiterate the article's three or four key phrases. ^{28, 41} They should add at least three or four more, and where more than one phrase or abbreviation is often used for the same thing, such as the proprietary and generic names of a drug, they should include every variant. ⁴¹ Keywords play a reduced role in search ranking but still have influence. ³³