Improving the quality of participant information sheets

new hot off the pressResearchers at Kings College London have just produced an interesting paper, published Feb 2016.  It is available online, and can be easily accessed. It looks at the ease of reading participant information sheets over a 10 year period – June 2003 to October 2013. Why is this interesting? We think it is an important piece of research and it is something the McPin Foundation has been getting more involved with of late. We have been helping study teams write plain English ‘lay’ summaries of their research ideas.  We have been trying to make information sheets more accessible to people they would like to recruit as participants in their studies.

An information sheet is very important in the conduct of research delivered to high ethical standards. Having an easy to understand information sheet is essential so that people can give ‘informed consent’. In this context, informed consent means having enough information to make a well thought out decision about whether you want to take part in a research study or not. An information sheet contains information such as:

  • Why are you asking me?
  • What do I have to do?
  • Will I or other people benefit?
  • Will any harm be done to me?
  • Who will know I took part?
  • How will my responses be stored and used?
  • Will my responses be confidential?

There is guidance on how to write an information sheet from bodies such as the Health Research Authority (HRA) and how to specifically write using plain English. But it isn’t easy and many research teams insert legal text, at the request of their own organisations, which is extremely hard to provide jargon free.

So what did the researchers Liam Ennis and Til Wykes find?  Having looked at 522 unique information sheets from studies within the National Institute of Health Research (NIHR) mental health portfolio data base they found some long and very complicated text.

  • For device intervention or pharmacological studies the average word count was 3000 – that is likely to run to 7 pages of text. The most complicated section of text in these sheets was graded “very difficult”. This needs to be seen in context. Many organisations insist that researchers use long and legalistic templates to base their information sheets on and legal departments of funders often include page upon page of text in an attempt to protect them from any possible legal action.
  • Studies that were working in the area of intellectual disability or dementia had information sheets that were easier to read. However the reading age grade was still 13-14 years (the UK national reading age).
  • Information sheets were shortest in studies based on observation. These did not need to explain complicated interventions or the process of randomisation.
  • One factor that seemed to reduce the length of information sheets was the involvement of patients in study design and delivery. They found as patient involvement increased in a study, the length of the sheet fell by about 150 words per level. However, patient involvement did not impact on readability.
  • They found that over time information sheets were getting longer in terms of word count.
    • In 2003, across all studies average word count was 1333 and by 2013 it was 1714.
  • They were not getting more complicated over time, but as already stated their readability as rated by reading age and a score known as the Flesch Index was high – average reading age 15-16 years and most complicated sections within the information sheet was measured as average age 18-19 years.
  • Ease of reading did not contribute to recruitment success in the studies assessed. Two factors that did influence recruitment. The first was level of patient involvement; the higher the involvement the better the recruitment to target. The second was intervention type, with pharmacological or devise studies less likely to recruit to time and target than other study designs.

What do we learn from this research? Firstly, it confirms our concerns at the McPin Foundation that often participant information sheets are overly complicated and hard to understand. Do people know what they are consenting to do? The data suggests that ease of reading information sheets does not impact on recruitment, but it may introduce bias into research samples. Are those with the stamina to read long information sheets more likely to enter studies than other people who might struggle to understand complex information? We know when your mental health is challenged, for many people concentrating and reading anything is difficult.

Secondly, the study reinforces the importance of involving patients in the design and delivery of studies as collaborators. More can be done to improve the use of plain English within information sheets, but it does seem that involving experts by experience keeps the lengths of these vital recruitment tools in check and helps ensure studies recruit to target. Writing in this way is a skill to be nurtured and developed, and we encourage the training of more service users and carers mental health research writers.

What else can we do to improve information sheets? Ennis and Wykes’ article suggests using a glossary to help make language more accessible and they provide a really helpful appendix to show how complicated text can be simplified and word length reduced without sacrificing meaning. We need informed research participants, who know exactly what is being asked of them, and for how long, so that retention within studies is high. The quality of a study is diminished if you lose lots of people during the course of the research, and thus have lots of missing data when you come to your analysis.

We have also started to think of ways of making information sheets better.

Two of the studies we are currently working on are producing a one page summary of the full information sheet as an easy read version. This gives people a choice. You can read the full version by yourself or you can read a summary and ask the research team to help you understand the full information sheet prior to giving informed consent. Having someone help you go through the full information sheet can make the whole process feel less daunting, and you can ask questions more easily. What is very important is the informed consent process can only take place after the full information sheet has been read and time provided for the person to make a decision over participation to prevent coercion.

Another idea we are progressing is to create an audio recording of the information sheet, or a video version. This would help people who find reading hard. The patient information sheet could be delivered in another format.

Our collective challenge is to reduce the average reading age of all mental health study information sheets to 13-14 years, the UK national average. We need to reduce their length and find a way that Ethics Committees, and research organisations, can balance their legal and ethical duties. People must have access to information to make an informed decision, but the length and complexity of information sheets should not be a barrier to engagement in research. Such barriers reinforce the problem that research study participants are not like “real world” people, and study findings that do not transition well from the research clinic to front line service delivery.

We tested this blog in terms of readability. The target is a score of 60 or more to meet the national average reading age standard of ‘plain English’.  We achieved a Flesch score of 54.8 which is in the fairly difficult to read band. We have more work to do improving the accessibility of our blog pieces.