How can information leaflets for medicines be made more readable?
Pharmaceutical information inserts can be a crucial factor in leading people to the correct or wrong use of medicines. More than thirty years ago, social scientists and critics of the pharmaceutical industry began to study the contents of these leaflets and compared information provided for the same product by the same company in different countries. They found that indications for use were limited in countries of ‘the North’ and numerous in ‘the South’. For contra indications, however, the opposite applied (see for example Silverman 1977). Since then, under public pressure, pharmaceutical companies have mended their ways to some extent, but information leaflets remain crucial and deserve continual vigilance. This article also draws attention to information leaflets. However, it does not focus on intentional misinformation in leaflets but on clarity and understandability of the instructions.
The European Commission (2006) summarises the quality requirements for Package Leaflets of medicinal products for human use as follows:
All medicines are required by Community law to be accompanied by outer/inner labelling text and a Package Leaflet setting out comprehensive information which is accessible to and understandable by those who receive it, so that they can use their medicine safely and appropriately. The safe and correct use of all medicines depends on users reading the labelling and packaging carefully and accurately and being able to understand and act on the information presented.
The European Commission Guideline focuses on readability and understandability of the leaflets from the patient’s point of view: accessibility and understandability for a safe and correct use. Indeed, for the patient it is essential to understand how and when he or she should take the appropriate medicine.
Furthermore, self-medication is increasing in Europe. Since self-medication implies consuming over-the-counter drugs, the patient does not necessarily ask for pharmaceutical advice. The only information patients get is provided by the Patient Information Leaflet. Surveys such as Patients’ Use and Perception of Medication Information Leaflets (Nathan 2007) show that 66% of patients do read the information leaflets, but no one knows how many patients understand them. How can documentation professionals improve the leaflets’ understandability and provide information tailored to the patient’s needs?
Evaluating understandability and readability
To provide accessible information for a safe and correct use of medicine, one should first consider understandability and readability.
For a text to score a high level of understandability, it should be understood, therefore made usable by the intended audience, namely the patient. Indeed, “Understandability is an ultimate prerequisite in order for users to perform any task” (http://tinyurl.com/3usegby). For patients, performing a task means deciding when and how to take a medicine in a safe way.
Taking a medicine in a safe way means that the patient complies with the recommendations provided by healthcare professionals and follows the instructions given in the leaflet. The final objective is to prevent any misuse or overdose of the medicine.
Government agencies, together with the pharmaceutical industry, confirm these requirements on how to use medicine in a safe way. For example, the PhRMa group (2011) reminds clients:
Taking your medicine the right way is important to make sure it works best for you:
- read the label
- don’t skip doses
- keep your medicines in the containers that they came in and do not remove the labels.
When stating “read the label”, the pharmaceutical industry implies “read and understand the leaflet.” Indeed, the US Federal Drug Agency, Consumer Affairs Office, published a brochure named “Use medicines safely” (http://tinyurl.com/89fd3tj) that recommends:
… understand how and when to take your medications (how many times a day and at what times, with or without food). Know if there are any special instructions (for example, shaking a liquid medication before using).
The danger of misuse is mentioned by the US National Institute of Aging available at http://tinyurl.com/6rzsa7w:
- Read the label—even if you have used the OTC product in the past. Important information can change.
- Remember, medicines—whether prescription or over-the-counter—can hurt you if they aren’t used the right way. Learn to be a smart consumer of medicine.
Readability is the ease with which a text can be read and understood. To measure readability, various indexes are available which aim at measuring how difficult a text may be for its readers. The method is based on sentence length and word length (Ganier 2002):
La plupart des formules de lisibilité…combinent deux aspects d’un texte pour prédire sa difficulté : la longueur des phrases, estimée par le nombre de mots qu’elles contiennent, et la complexité des mots, estimée par le nombre de syllabes.
While several readability indexes are available and commonly used, their efficiency is being questioned during Patient Information Leaflet tests and by experts (Ganier 2002, Redish 1981).
For the English language, the Flesch Reading Easeindexis commonly used. It is, indeed, based on word length and sentence length: the longer the word or the longer the sentence, the more difficult the text. For the French language, practitioners use the Kandel and Moles readability test. As described by the McGill University Health Center in the Chapter “MUHC Tool Kit for creating Patient Learning Materials” (http://tinyurl.com/79zuh87),
… the Kandel & Moles test rates a document on a 100 point scale. The higher the score, the easier the document is to read. Most documents for average adult readers should have a score between 60 and 70, which roughly mirrors a grade 6 to 8 reading level.
The Spanish readability test was published by Fernández Huerta and is available at http://www.standards-schmandards.com/exhibits/rix/ . The Huerta Reading Index was analysed by the University of Texas at Austin, in conjunction with the Texas Technology Access Program/Texas Center for Disability Studies (http://tinyurl.com/d3b8pgk). They conclude:
[The Huerta Reading Index] requires the counts of the average number of syllables and sentences to be done in 100-word chunks of text. Additionally, the formula relies on sentence-ending punctuation to determine the number of words used in analysis and where to demarcate chunks of text. The Huerta Reading Index analysis reports the total number of words submitted but only analyzes the words within sentences.
A first test of a French sentence (a) gives an insight of the applicability of such indices: “Ce médicament contient de l‘aspirine, mais ce dosage est plus particulièrement destiné au traitement de certaines affections du cœur ou des vaisseaux, seul ou associé aux autres traitements prescrits par votre médecin.” This text scores 34, which denotes a difficult sentence.
The second sample (b): “L‘aspirine augmente les risques hémorragiques et ceci dès les très faibles doses et même lorsque la prise date de plusieurs jours” scores 56, which is graded as ‘fairly difficult’. The readability index rates both samples as ‘difficult/fairly difficult’, still, each sample is based on a single sentence, made up of commonly used words. There are no complicated medical terms in these descriptions. Both sentences are grammatically correct and constructed on the plain Subject-Verb-Object (SVO) pattern. Sample (a) has 32 words, whereas sample (b) has 21 words; the sentence length is immediately reflected in the readability score: the longer the sentence, the more difficult it is to understand.
Furthermore, the word length is relevant here. Sample (a) contains two so-called ‘long words’: médicament,particulièrement. Sample (b) contains a single ‘long word’: Hémorragique. This shows the limitation of such indexes: a word is considered long if it has more than three syllables. In languages such as French, this size limit is inappropriate. For example, the word ‘excipients’ should be considered as ‘easy-to-understand’, whereas ‘électrocardiogramme’ is evaluated as ‘long and complicated’. This classification does not reflect the patient reality. Whereas ‘électrocardiogramme’ is well-understood because familiar, ‘excipients’ is a highly medical term that, in a documentation dedicated to patient information, should be replaced by ‘composants’ in French. This phenomenon is confirmed (Ganier 2002) as
“…certain mots pollysyllabiques longs (canalisation, commutateur, imprimante) peuvent affecter négativement les scores des formules de lisibilités basées sur le comptage du nombre de syllables… sans toutefois perturber la compréhension de l’utilisateur.”
For the English language, Redish (1981) stresses another limitation of such indexes:
A formula that counts only sentence length and word length or familiarity of the words is not sensitive to the order of the words or the complexity of the grammar. Sentences with misplaced clauses, dangling participles, or misused words will score as well on a readability formula as sentences of equal length that have none of these problems.
Indexes are not sensitive to the order of the words. Moving words within sentences – and thus producing incorrect sentences – has no impact on the readability index.
User testing (evaluating understandability)
Interviewing test persons is described in the September 2006 revision of the European Commission Guideline on Readability of Package Leaflet. This process is known as “User testing” and aims at “indicating where there are problem areas which should be rectified.”
The European Directive 2004/27/EC states:“The package leaflet shall reflect the results of consultations with target patient groups to ensure that it is legible, clear and easy to use.” The associated document‘Guidance concerning consultations with target patient groups for the package leaflet’ says:
One of the possible ways of complying with the legal requirement is by performing a ‘user testing’ of the package leaflet. ‘User testing’ means to test the readability of a specimen with a group of selected test subjects. It is a development tool which is flexible and aims to identify whether or not the information as presented, conveys the correct messages to those who read it. Testing itself does not improve the quality of the information but it will indicate where there are problems areas.
The essential rule defined in the September 2006 European Commission Guideline reads: “Information which can be used by the least able will be beneficial for all users.”
Indeed, it is essential to test Patient Information Leaflets with people who have little experience with medicine and who “find written information difficult and… do not use written documents in their working life.” Collecting user tests from a minimum of 20 people renders a reasonable insight in the understandability of a Patient Information leaflet and is more reliable than a readability index.
Apart from measuring readability via readability indices or patient interviews, a greater success in user-friendliness might be obtained by improving the Patient Information Leaflet design from the very beginning. “The design and layout of the information is crucial in helping patients to find and understand the important messages within the PIL” stresses the Medical Health products Regulatory Agency (MHRA 2011). The British Medical Health products Regulatory Agency insists the lay-out should help the patient to “navigate the information.” Lay-out parameters include headings, colours, text size, white space, bulleted lists, grouping side effects, avoiding repeating information (confusing) and eliminating boxes (overlooked by patients).
Combining the Medical Health products Regulatory Agency recommendations with Sless’ (2006) research studies and the September 2006 European Commission Guideline, it is recommended to organize the Patient Information Leaflet template as follows. First, it is essential to reduce the information volume, known as information overdose: cut, re-organize, minimalize are the basic rules. Second, all similar information should be grouped under a clearly defined topic. Then, information should be structured. For example:
1. Before taking the drug
1.1. Don‘t take the medicine if…
1.2. If you take other medicines
1.3. If you are pregnant
1.4. If you are driving
2. How to take this drug
2.1 Check the dose
2.2. Check the maximum duration of treatment
Grouping the information prevents repeating data, which is confusing for the patient. Having the same information repeated up to three times in a single Patient Information Leaflet either makes patients anxious about their capacity to understand or leads them to reject such instructions as insulting to their intelligence.
Limiting the word length might be a point to consider, although such rules are not applicable to Romanic languages or to German or Finnish, for example. Screening a text for highly technical (medical) terms might also be advisable. Instead of ‘posology’, patient leaflets should use a more familiar word such as ‘dosage. Similarly, ‘components’ belongs to patient daily vocabulary and should be used instead of ‘excipients’. Avoiding synonyms helps reduce patient confusion. Using ‘components’ and ‘constituents’ in the same leaflet does not help users understand that these words cover the same medical meaning, particularly if they were faced with the word ‘excipients’ in another Patient Information Leaflet.
The September 2006 European Commission Guideline recommends limiting the sentence length to 20 words in English. Professional writers agree upon a maximum length of 27 words in French. Short sentences definitely help users maintain their attention and quickly grasp the meaning of the leaflet recommendations.
Maintaining a user-friendly style that makes great use of action verbs, active voice and parallelism guarantees a high level of understanding to all patients. Parallelism (repeating a sentence structure as often as possible in the leaflet text) is recommended.
Typography and lay-out
Texts written in all capital letters should be avoided according to the September 2006 European Commission Guideline. ALL CAPS definitely reduces the legibility level of a text. Further to the most recent eye-tracking surveys, using centred paragraphs should be abandoned, as centred text has a negative impact on legibility. Displaying information in outline form, instead of block text, and using bullet points has proven to be more appealing, easier to process and more effective than other lay-outs. It also helps maintain attention longer (Wogalter 1999).
Warning a patient of a potential risk does not mean pasting multiple safety messages in the package leaflet. Users do not read warnings; users need plain, understandable information. Shrensky (2011) demonstrates how useless framed warnings are: “put simply, people avoid reading things in boxes because they ignore anything that looks like an advertisement.” Additionally, Wogalter (1999: 127), the ‘Warnings Expert’, concludes that “the presence of a border slowed performance in a rapid recognition task.”
Symbols, icons and pictograms
‘A picture is worth a thousand words’ is not always applicable in the field of package leaflets. Showing Figure 1 on a package leaflet does not bring the expected result. In testing how women interpreted this pregnant woman symbol, the US Food and Drug Administration observed that only 21 in 100 women knew the meaning of the symbol. More astonishing was the fact that only 27 in 100 women thought the symbol meant that the package contained birth control medicine. During [Product name’s] first ten years on the market, there were 2,181 reports of pregnancy while taking the medicine. Most women had abortions, but almost half of the babies that were born had birth defects (FDA Consumer 22/8: 26).
In Healthcare documentation, the time has come to break the myth of self-explanatory symbols. The Children’s Hospital of Pittsburgh (USA) analysed how children understood the skull and crossbones pictogram (Fig. 2). While adults immediately think “Beware! Poison”, children react: “Let’s play pirates” (Wogalter 1999: 176). This ambiguity is clearly stressed by the European Medicine Agency (EMEA 2007, section B., p. 2): “The main risk is misunderstanding of the symbol… Misinterpretation by the public (families, older children themselves, carers) has led, can and will lead to medication errors.”
Restricting the use of pictograms is recommended by the European Commission Guideline (2006, section 8): “Symbols and pictograms… should only be used to aid navigation, clarify or highlight certain aspects of the text and should not replace the actual text.” Teams might also be tempted to create their own symbols. Green (2011) reports on his experience with students designing symbols. Each student presented his symbols to class to test comprehension. The success rate was very low (20% at best). This is well below the ANSI (American National Standard Institute) recommendation of 85% recognition in order to be validated.
Additionally, the European Commission Guideline (2006, section 8) points out the problem of language and cultural acceptance: “Particular care will be needed when symbols are transferred or used in other language versions of the leaflet and further user testing of these may be necessary.”Indeed, instead of spending time and energy illustrating the package leaflet by means of boxed texts or pictograms, authors would be well-advised to focus on simplicity, clarity and user-testing.
Utilising readability tests with caution is a first step in improving the understandability of Patient Information Leaflets. Eliminating useless or misleading information, structuring text, and using easily understood terminology is a second step in providing patients with useful and understandable package leaflets. Finally, applying basic technical communication rules - i.e. putting the patient in the centre of the document design process - combined with user tests will help ensure a user-friendly Patient Information Leaflet which will enhance proper medicine use.
Marie-Louise Flacke is a graduate from the American University of Paris (Technical Writing Program). From 2003-2006, she was vice-president of INTECOM (worldwide association of Technical Communication professionals). In 2007, she set-up the Technical Writing Program at the Université de Haute-Bretagne (Rennes 2, France) and has been training Master2 students at the Université de Bretagne Occidentale (Brest, France). Additionally, she specialised in designing user-friendly documentation in the medical device field. Her most recent research focused on “Handling warnings in user-friendly documentation” and “Iconmania: don’t ruin your documentation efforts!” Email:firstname.lastname@example.org
EMEA (European Medicines Agency)
2007 Pre-authorisation evaluation of medicines for human use. Recommendation of the Paediatric Committee to the European Commission regarding the symbol. Ref. EMEA/498247/2007
2006 Guideline on the readability of the label and package leaflet of medicinal products for human use.The European Commission, Enterprise and Industry Directorate-General, Directive 2011/83/EC, Revision September 2006.
2002 Evaluer l'effacité des documents techniques procéduraux: Un panorama des méthodes. Le Travail Humain 65 (1): 1-2.
2011 Warnings and warning labels. In:www.visualexpert.com/Resources/warnings.html
(accessed September 2011).
MHRA (Medical Health products Regulatory Agency)
2011 Further guidance on designing Patient Information Leaflets and how to achieve success in User Testing. In: http://www.cambreg.co.uk/_downloads/Ref6-Further-Guidance.pdf (accessed September 2011).
2007 Patients' Use and Perception of Medication Information Leaflets The Annals of Pharmacotherapy May 2007 vol. 41 no. 5 777-782
2011 Taking medicines safely
(Accessed November 2011)
1981 Understanding the limitations of readability formulas. IEEE Transactions on Professional Communication PC-24 (1): 46-48.
2011 Boxed risk warnings: Research findings. Melbourne:Communication Research Institute.http://communication.org.au/publications/reviews/Boxed-risk-warnings--research-findings/67,32.html (accessed September 2011).
1977 The epidemiology of drug promotion. International Journal of Health Services 7 (2): 157-66.
2006 Writing about medicine for people. Communication Research Institute, Melbourne, Australia
1999 Warnings and risk communication, London: Taylor & Francis.