- Email: [email protected]
Evidence That Patients can Understand and Use?
ARTICLE IN PRESS
www.elsevier.de/zefq Z. Evid. Fortbild. Qual. Gesundh. wesen (ZEFQ) 102 (2008) 493–496
Schwerpunkt
Evidence That Patients can Understand and Use? Nancy Santesso1,, Claire Glenton2, Britta Lang3 1
IFO Italian National Cancer Institute Regina Elena, Rome, Italy Norwegian Knowledge Centre for the Health Services, Oslo, Norway 3 German Cochrane Centre, Freiburg Germany 2
Summary The impact that research evidence can have on patients making decisions about their own care is often overlooked. Regardless of whether patients want to lead or fully participate in their care, it appears that patients still want to be informed. We have health information that patients could benefit from knowing, a growing body of evidence and experiences about how to translate this information to patients. The Cochrane Collaboration, an organisation that produces and disseminates systematic reviews of healthcare interventions, translates the evidence from their reviews into summaries for the public. Research is presently being conducted to test a
format for the plain language summaries. Principles such as presenting evidence in standard qualitative statements, as relative and absolute risks, as natural frequencies, and in tables are being incorporated into the format. Groups within the Collaboration are also translating the evidence for use by patients and the public. Effective strategies are being used to target patients directly and via their physicians, such as dissemination on specific patient group websites and decision aids. Research evidence and experiences can feed into the creation of evidence that patients can understand and use.
Key words: patients, patient participation, consumer health information, Cochrane Collaboration
Evidenz, die Patienten verstehen und nutzen ko¨nnen? Zusammenfassung Ha¨ufig wird u¨bersehen, welchen Einfluss wissenschaftliche Evidenz auf die Entscheidungen haben kann, die Patienten hinsichtlich ihrer medizinischen Versorgung treffen. Unabha¨ngig davon, ob Patienten ihre Behandlung maßgeblich mitbestimmen oder an den Entscheidungen daru¨ber voll partizipieren, wollen sie anscheinend informiert werden. Wir verfu¨gen u¨ber Gesundheitsinformationen, von denen die Patienten profitieren ko¨nnten, sowie eine zunehmende Evidenzbasis und die entsprechenden Erfahrungen, wie man diese Informationen fu¨r Patienten aufbereitet. Die Cochrane Collaboration erstellt und verbreitet systematische Reviews zu gesundheitlichen Interventionen und u¨bersetzt die Evidenz aus ihren ’’
Reviews in Zusammenfassungen, die fu¨r die O¨ffentlichkeit versta¨ndlich sind. Ein Format fu¨r die Zusammenfassung in allgemein versta¨ndlicher Sprache wird derzeit im Rahmen einer wissenschaftlichen Untersuchung getestet. Dieses Format umfasst bestimmte Prinzipien wie die Darstellung von Evidenz in Form standardisierter qualitativer Aussagen, als relative und absolute Risiken, als natu¨rliche Ha¨ufigkeiten und in Tabellenform. Verschiedene Gruppen innerhalb der Cochrane Collaboration bereiten die Evidenz auch fu¨r Patienten und fu¨r die O¨ffentlichkeit auf. Um die Patienten direkt und u¨ber ihre A¨rzte anzusprechen, kommen effektive Strategien zum Einsatz, etwa die Verbreitung auf den Internetseiten
’’
Corresponding author. Nancy Santesso, S.C. Epidemiology, IFO Italian National Cancer Institute Regina Elena, Via Elio Chianesi 53, Rome – Italy 00144.
Tel.: +39 06 5266 2762; fax: +39 06 5266 2732. E-Mail: [email protected] (N. Santesso). Z. Evid. Fortbild. Qual. Gesundh. wesen (ZEFQ) doi:10.1016/j.zefq.2008.08.027
493
ARTICLE IN PRESS spezieller Patientengruppen und Entscheidungshilfen. So ko¨nnen wissenschaftliche Evidenz und Erfahrungen in die Erstellung von Informationen
einfließen, die die Patienten auch verstehen und anwenden ko¨nnen.
Schlu¨sselwo¨rter: Patienten, Patientenbeteiligung, Patienteninformationen, Cochrane Collaboration
Evidence that Patients can Understand and Use? Traditionally health research evidence has been written for researchers and health care professionals with an increasing interest in policy makers and other decision makers. Perhaps viewed as less noble or valuable, is targeting patients and consumers with this information. But consider the decisions that patients make everyday about their health. They make decisions about whether they should visit their doctor, whether they ask their doctors to give them antibiotics for a cold, whether they actually take their prescriptions or follow other advice from their doctors, or whether they apply pressure on policy makers to approve or add a drug or vaccine to a national formulary. But the impact that research evidence can have on patients making decisions about their own care is often overlooked. There is a great potential in translating high quality evidence into information that patients can understand and use, and in reaching patients with that information. At this point, we have health information that patients could benefit from know-
ing, we have a growing body of evidence about how to translate that research information into a form patients can understand, and we have first-hand experience about how to do it, in particular over 15 years of experiences in the Cochrane Collaboration.
But why provide patients with information? Regardless of whether patients want to lead or fully participate in their care or whether they want their doctors to make their health care decisions for them, it appears that patients still want to be informed. In a survey conducted in the UK with 3000 people over the age of 45, four out of five people said they would likely seek information to cope with their health concerns [1]. The survey also asked where and from whom they looked for health information. Nearly three quarters said that they turned to their doctors, while one fifth referred to other health care professionals. The second most commonly cited source of health information was the Internet, referred to by a third (see Fig. 1). But is this just a UK phenomenon?
A systematic review of information needs among cancer patients, with studies from Europe and Canada, also points to doctors as the main information source [2]. Notable is that even in this review of studies from 1980 to 2003 the Internet was already one of the top three sources. With increasing access to the Internet, the Internet may increasingly become the principle source of information. In Germany, for instance, over half the population has Internet access and this proportion is predicted to increase [3]. One of the most worrisome issues with this access is the fear that information on the Internet may do more harm than good. Are the stories about patients who die prematurely because they stopped chemotherapy treatment to instead take a series of holistic supplements advertised on the web urban myths? Or are these stories a consequence of information on the Internet causing harm? Is it true that the Internet does more harm than good? A review by Eysenbach [4] in 2002, reports that we cannot answer this question, but that it is possible to identify some of the issues and perhaps solve them. Eysenbach’s main conclusion is that the proportion of bad or good quality information may be the same on the Internet as it is for other media, including magazines. However, the risk that people will find bad quality information on the Internet could be greater because there is more information overall on the Internet and it is more easily accessible. The solution to this dilemma is to provide ‘‘safe’’ web sites that provide filtered and reliable information.
How has the research community responded?
Fig. 1. Sources of health information from the Picker Institute Europe Survey [1].
494
Research organisations and funding bodies now create their own ‘‘safe sites’’ with high quality research evidence for patients. BMJ Best Treatments Z. Evid. Fortbild. Qual. Gesundh. wesen 102 (2008) 493–496 www.elsevier.de/zefq
ARTICLE IN PRESS (http://www.besttreatments.org), for example, is a web site with information that has been translated from Clinical Evidence into a user friendly format; NICE (http://www.nice.org.uk/), an organisation in the UK that produces guidance based on the best available evidence, translates information from their guidance documents into documents ‘‘for patients, carers and the public’’; and the Journal of the American Medical Association (http://jama. ama-assn.org/cgi/collection/patient_page) includes in their journal ‘‘Patient Pages’’ which are summaries of evidence from their publications. The Cochrane Collaboration, an organisation that produces, disseminates and updates systematic reviews of healthcare interventions, also translates evidence from their reviews into summaries for the public. It has produced these summaries since its inception over 15 years ago, and has made them freely available on the Internet (http://www.cochrane.org/reviews/). Each systematic review should include a synopsis or plain language summary which summarises the review in a ‘‘straightforward style that can be understood by consumers of health care’’ [5]. Guidelines on how to write the summary are provided in the Cochrane Handbook, and the summary is peer reviewed during the review process. Until now the guidelines have been quite general, with few restrictions on content and presentation of information for consumers. With over 10 000 authors contributing to and writing these summaries, are the summaries good? In 2004, a group from the German Cochrane Centre reviewed the quality of the plain language summaries [6]. They concluded that ‘‘a substantial number of synopses have errors and some do not have a sufficient quality for publication’’. The study analysed 243 randomly sampled plain language summaries in the Cochrane Library. The authors compared the summaries with the abstracts (and if necessary, the review) and found that 24% of the summaries and abstracts did not match. Errors included factual errors, discrepancies in stated facts between the summary and the review, discrepancies in the strength of
evidence claimed between the summary and the review, and common wording errors. On the other hand, 76% of the summaries did not contain errors.
Do patients understand ‘plain language summaries’? What is not known, however, is whether people understand the summaries. Some work has been done by a few Cochrane Review Groups to assess the summaries. For example, the Norwegian branch of the Nordic Cochrane Centre has evaluated summaries of Cochrane reviews written for back pain sufferers [7] and the Cochrane Musculoskeletal Group has completed user testing with summaries for people with arthritis [8]. Based on that work and the work carried out by the German Cochrane Centre, it has become clear that the Cochrane Collaboration needs to develop better guidelines for authors writing plain language summaries. During 2008, with the support of the Cochrane Collaboration’s Opportunity Fund, six Cochrane entities will develop guidelines and formats for plain language summaries and test these among consumers and consumer representatives. This work will build upon a variety of experiences and research: the work carried out by the Norwegian Branch and the Musculoskeletal Group; guidelines from other organisations for writing patient information and research evidence for the presentation of information to patients and the public. Several issues for which there is evidence and other issues that have not yet been evaluated will be addressed during this work. For example, in current plain language summaries authors use a variety of words to express the strength of the evidence and the magnitude of the effect of the interventions. There is a need to develop standard qualitative phrases to avoid phrases such asy.’’a moderately large effect’’ or ‘‘a high likelihood of somewhat small but possibly important effects’’. Qualitative phrases do have shortcomings since they are interpreted differently by dif-
Z. Evid. Fortbild. Qual. Gesundh. wesen 102 (2008) 493–496 www.elsevier.de/zefq
ferent people [9]. But standard phrases, used by authors, consistently may decrease misinterpretation. In contrast, quantitative statements, that is, providing numbers to describe the evidence, give a more precise estimate of effect. Presenting numbers, however, has its own challenges. But despite the challenges, there are a few principles that can be incorporated into patient information. Natural frequencies are preferred (e.g. 5 out of 100). When expressing benefits or risks, it is better to not just describe the relative risks alone, but to include absolute risks, and to compare what happens to people with or without the intervention (e.g. 5 out of 100 had outcome without Drug A, 20 out of 100 had outcome with Drug A) [9]. There is also some evidence that people prefer to read information in tables [10]. Other issues, not yet evaluated will also be investigated when testing the various formats, such as how to present the quality of evidence and, if helpful, how to present confidence intervals, and how much information to give about the research that lies behind these results. Using these principles, a variety of potential formats for the plain language summaries have been developed. These formats will be tested in different countries and among different groups. This research should potentially be useful not only for the Cochrane Collaboration, but also for other groups preparing patient information. Once written, what more can Cochrane do? What more can other organisations do? The next step is to ensure that the information is used. Although a relatively new area of research, there is some evidence about how to promote the use of research information by patients. Cochrane and a number of Cochrane groups are already working on this knowledge translation to consumers.
How can we promote the use of plain language summaries? One of the simplest strategies, of course, is taking advantage of the
495
ARTICLE IN PRESS number of people accessing the Internet and seeking health information. As previously described, the Cochrane Collaboration has created a ‘safe site’ to search for and read the plain language summaries of Cochrane reviews. Many Cochrane Review groups also link to or have packaged these summaries on their own web sites for easy access (e.g. the Cochrane Wounds and the Cochrane Neonatal Groups) or produced separate patient information web sites (e.g. the Back Review Group www.rygginfo.no). Instead of creating a new site to house the plain language summaries, the Cochrane Musculoskeletal Group, for example, has taken knowledge translation one step further and provides its summaries on a variety of patient group web sites to directly reach patients and consumers. This group has been working with success with a couple of arthritis groups: The Arthritis Society in Canada (www. arthritis.ca/look at research/cochrane reviews and Arthritis Victoria in Melbourne (www.arthritisvic.org.au)). The summaries are stand-alone products on these web sites but are often incorporated into the general patient information provided on the site. Even as stand-alone products consumers visiting the Arthritis Society web site indicated that the summaries increased their knowledge and were useful for decision making [8]. Since nearly three quarters of people seek information from their physicians, targeting patients via their physicians is another strategy for knowledge translation. Evidence shows that patients benefit more from written information when used in consultation with health care professionals [11]. Research organisations who, as described previously, produce patient information alongside information for professionals, may increase the likelihood that physicians see patient information and use it in consultations. Decision aids, which present the information from the plain language summaries with tools to help
496
patients make health care decisions with their physician, have also proven to be an effective knowledge translation strategy. Decision aids have been shown to improve patient knowledge, and align the decisions that patients make with their values and expectations [12]. The Cochrane Musculoskeletal Group has produced and continues to produce decision aids based on their reviews and more research to evaluate their effectiveness is pending [8].
[2]
[3]
[4]
So, can we do it? Ultimately, the purpose of any translation of health research knowledge into formats that researchers, health care professionals, policy makers and patients can understand and use is to improve patient care. Informing patients directly makes sense, but it can be challenging. There is already some research evidence about how to make research information patientfriendly, but more needs to be done. There is also some evidence about strategies to ensure that the information is used. While it is important to build on and use this evidence, it is also valuable to learn from the experiences of others. The Cochrane Collaboration has over 15 years of experience incorporating the principles of how to present research to patients and facilitate its use. Therefore, we hope our experiences can feed into the existing research evidence, as well as into the efforts of other research organisations currently working in this field. In this way, we may create ‘evidence that patients can understand and use’. Funding: Cochrane Collaboration Opportunities Fund Canadian Institutes of Health Research
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
References [1] Ellins J, Coulter A. How engaged are people in their health care? Findings of a
national telephone survey. Oxford: Picker Institute Europe; 2005. Rutten LJ, Arora NK, Bakos AD, Aziz N, Rowland J. Information needs and sources of information among cancer patients: a systematic review of research (1980–2003). Patient Educ Couns 2005 Jun;57(3):250–61. N(o)nliner-Atlas 2007. TNS infratest, 2007. Available from /http://www.initiatived21. de/fileadmin/files/NOA_Umzug/NOA_ Atlanten/NONLINER-Atlas2007.pdfS. Eysenbach G, Powell J, Kuss O, Sa ER. Empirical studies assessing the quality of health information for consumers on the world wide web: a systematic review. JAMA 2002 May 22–29;287(20): 2691–700. Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.0.0 [updated February 2008]. The Cochrane Collaboration, 2008. Available from /http://www.cochranehandbook.orgS. Falck-Ytter C, Lang B, Booker D, Antes G, Falck-Ytter Y. The quality of Cochrane synopses – a quality assessment pilot project. 12th Cochrane Colloquium. Ottawa, 2004. Available from /http:// www.cochrane.org/colloquia/abstracts/ ottawa/P-040.htmS. Glenton C, Nilsen ES, Carlsen B. Lay perceptions of evidence-based information – a qualitative evaluation of a website for back pain sufferers. BMC Health Serv Res 2006 Mar 15;6:34. Tugwell PS, Santesso NA, O’Connor AM, Wilson AJ, Effective Consumer Investigative Group. Knowledge translation for effective consumers. Phys Ther 2007 Dec;87(12):1728–38. Lipkus IM. Numeric, verbal, and visual formats of conveying health risks: suggested best practices and future recommendations. Med Decis Making 2007 Sep–Oct;27(5):696–713. Schu¨nemann HJ, Sta( hl E, Austin P, Akl E, Armstrong D, Guyatt GH. A comparison of narrative and table formats for presenting hypothetical health states to patients with gastrointestinal or pulmonary disease. Med Decis Making 2004 Jan–Feb;24(1):53–60. Coulter A, Ellins J. Effectiveness of strategies for informing, educating, and involving patients. BMJ 2007 Jul 7;335(7609): 24–7. O’Connor AM, Stacey D, Entwistle V, Llewellyn-Thomas H, Rovner D, HolmesRovner M, et al. Decision aids for people facing health treatment or screening decisions. Cochrane Database Syst Rev 2003(2):CD001431.
Z. Evid. Fortbild. Qual. Gesundh. wesen 102 (2008) 493–496 www.elsevier.de/zefq
www.elsevier.de/zefq Z. Evid. Fortbild. Qual. Gesundh. wesen (ZEFQ) 102 (2008) 493–496
Schwerpunkt
Evidence That Patients can Understand and Use? Nancy Santesso1,, Claire Glenton2, Britta Lang3 1
IFO Italian National Cancer Institute Regina Elena, Rome, Italy Norwegian Knowledge Centre for the Health Services, Oslo, Norway 3 German Cochrane Centre, Freiburg Germany 2
Summary The impact that research evidence can have on patients making decisions about their own care is often overlooked. Regardless of whether patients want to lead or fully participate in their care, it appears that patients still want to be informed. We have health information that patients could benefit from knowing, a growing body of evidence and experiences about how to translate this information to patients. The Cochrane Collaboration, an organisation that produces and disseminates systematic reviews of healthcare interventions, translates the evidence from their reviews into summaries for the public. Research is presently being conducted to test a
format for the plain language summaries. Principles such as presenting evidence in standard qualitative statements, as relative and absolute risks, as natural frequencies, and in tables are being incorporated into the format. Groups within the Collaboration are also translating the evidence for use by patients and the public. Effective strategies are being used to target patients directly and via their physicians, such as dissemination on specific patient group websites and decision aids. Research evidence and experiences can feed into the creation of evidence that patients can understand and use.
Key words: patients, patient participation, consumer health information, Cochrane Collaboration
Evidenz, die Patienten verstehen und nutzen ko¨nnen? Zusammenfassung Ha¨ufig wird u¨bersehen, welchen Einfluss wissenschaftliche Evidenz auf die Entscheidungen haben kann, die Patienten hinsichtlich ihrer medizinischen Versorgung treffen. Unabha¨ngig davon, ob Patienten ihre Behandlung maßgeblich mitbestimmen oder an den Entscheidungen daru¨ber voll partizipieren, wollen sie anscheinend informiert werden. Wir verfu¨gen u¨ber Gesundheitsinformationen, von denen die Patienten profitieren ko¨nnten, sowie eine zunehmende Evidenzbasis und die entsprechenden Erfahrungen, wie man diese Informationen fu¨r Patienten aufbereitet. Die Cochrane Collaboration erstellt und verbreitet systematische Reviews zu gesundheitlichen Interventionen und u¨bersetzt die Evidenz aus ihren ’’
Reviews in Zusammenfassungen, die fu¨r die O¨ffentlichkeit versta¨ndlich sind. Ein Format fu¨r die Zusammenfassung in allgemein versta¨ndlicher Sprache wird derzeit im Rahmen einer wissenschaftlichen Untersuchung getestet. Dieses Format umfasst bestimmte Prinzipien wie die Darstellung von Evidenz in Form standardisierter qualitativer Aussagen, als relative und absolute Risiken, als natu¨rliche Ha¨ufigkeiten und in Tabellenform. Verschiedene Gruppen innerhalb der Cochrane Collaboration bereiten die Evidenz auch fu¨r Patienten und fu¨r die O¨ffentlichkeit auf. Um die Patienten direkt und u¨ber ihre A¨rzte anzusprechen, kommen effektive Strategien zum Einsatz, etwa die Verbreitung auf den Internetseiten
’’
Corresponding author. Nancy Santesso, S.C. Epidemiology, IFO Italian National Cancer Institute Regina Elena, Via Elio Chianesi 53, Rome – Italy 00144.
Tel.: +39 06 5266 2762; fax: +39 06 5266 2732. E-Mail: [email protected] (N. Santesso). Z. Evid. Fortbild. Qual. Gesundh. wesen (ZEFQ) doi:10.1016/j.zefq.2008.08.027
493
ARTICLE IN PRESS spezieller Patientengruppen und Entscheidungshilfen. So ko¨nnen wissenschaftliche Evidenz und Erfahrungen in die Erstellung von Informationen
einfließen, die die Patienten auch verstehen und anwenden ko¨nnen.
Schlu¨sselwo¨rter: Patienten, Patientenbeteiligung, Patienteninformationen, Cochrane Collaboration
Evidence that Patients can Understand and Use? Traditionally health research evidence has been written for researchers and health care professionals with an increasing interest in policy makers and other decision makers. Perhaps viewed as less noble or valuable, is targeting patients and consumers with this information. But consider the decisions that patients make everyday about their health. They make decisions about whether they should visit their doctor, whether they ask their doctors to give them antibiotics for a cold, whether they actually take their prescriptions or follow other advice from their doctors, or whether they apply pressure on policy makers to approve or add a drug or vaccine to a national formulary. But the impact that research evidence can have on patients making decisions about their own care is often overlooked. There is a great potential in translating high quality evidence into information that patients can understand and use, and in reaching patients with that information. At this point, we have health information that patients could benefit from know-
ing, we have a growing body of evidence about how to translate that research information into a form patients can understand, and we have first-hand experience about how to do it, in particular over 15 years of experiences in the Cochrane Collaboration.
But why provide patients with information? Regardless of whether patients want to lead or fully participate in their care or whether they want their doctors to make their health care decisions for them, it appears that patients still want to be informed. In a survey conducted in the UK with 3000 people over the age of 45, four out of five people said they would likely seek information to cope with their health concerns [1]. The survey also asked where and from whom they looked for health information. Nearly three quarters said that they turned to their doctors, while one fifth referred to other health care professionals. The second most commonly cited source of health information was the Internet, referred to by a third (see Fig. 1). But is this just a UK phenomenon?
A systematic review of information needs among cancer patients, with studies from Europe and Canada, also points to doctors as the main information source [2]. Notable is that even in this review of studies from 1980 to 2003 the Internet was already one of the top three sources. With increasing access to the Internet, the Internet may increasingly become the principle source of information. In Germany, for instance, over half the population has Internet access and this proportion is predicted to increase [3]. One of the most worrisome issues with this access is the fear that information on the Internet may do more harm than good. Are the stories about patients who die prematurely because they stopped chemotherapy treatment to instead take a series of holistic supplements advertised on the web urban myths? Or are these stories a consequence of information on the Internet causing harm? Is it true that the Internet does more harm than good? A review by Eysenbach [4] in 2002, reports that we cannot answer this question, but that it is possible to identify some of the issues and perhaps solve them. Eysenbach’s main conclusion is that the proportion of bad or good quality information may be the same on the Internet as it is for other media, including magazines. However, the risk that people will find bad quality information on the Internet could be greater because there is more information overall on the Internet and it is more easily accessible. The solution to this dilemma is to provide ‘‘safe’’ web sites that provide filtered and reliable information.
How has the research community responded?
Fig. 1. Sources of health information from the Picker Institute Europe Survey [1].
494
Research organisations and funding bodies now create their own ‘‘safe sites’’ with high quality research evidence for patients. BMJ Best Treatments Z. Evid. Fortbild. Qual. Gesundh. wesen 102 (2008) 493–496 www.elsevier.de/zefq
ARTICLE IN PRESS (http://www.besttreatments.org), for example, is a web site with information that has been translated from Clinical Evidence into a user friendly format; NICE (http://www.nice.org.uk/), an organisation in the UK that produces guidance based on the best available evidence, translates information from their guidance documents into documents ‘‘for patients, carers and the public’’; and the Journal of the American Medical Association (http://jama. ama-assn.org/cgi/collection/patient_page) includes in their journal ‘‘Patient Pages’’ which are summaries of evidence from their publications. The Cochrane Collaboration, an organisation that produces, disseminates and updates systematic reviews of healthcare interventions, also translates evidence from their reviews into summaries for the public. It has produced these summaries since its inception over 15 years ago, and has made them freely available on the Internet (http://www.cochrane.org/reviews/). Each systematic review should include a synopsis or plain language summary which summarises the review in a ‘‘straightforward style that can be understood by consumers of health care’’ [5]. Guidelines on how to write the summary are provided in the Cochrane Handbook, and the summary is peer reviewed during the review process. Until now the guidelines have been quite general, with few restrictions on content and presentation of information for consumers. With over 10 000 authors contributing to and writing these summaries, are the summaries good? In 2004, a group from the German Cochrane Centre reviewed the quality of the plain language summaries [6]. They concluded that ‘‘a substantial number of synopses have errors and some do not have a sufficient quality for publication’’. The study analysed 243 randomly sampled plain language summaries in the Cochrane Library. The authors compared the summaries with the abstracts (and if necessary, the review) and found that 24% of the summaries and abstracts did not match. Errors included factual errors, discrepancies in stated facts between the summary and the review, discrepancies in the strength of
evidence claimed between the summary and the review, and common wording errors. On the other hand, 76% of the summaries did not contain errors.
Do patients understand ‘plain language summaries’? What is not known, however, is whether people understand the summaries. Some work has been done by a few Cochrane Review Groups to assess the summaries. For example, the Norwegian branch of the Nordic Cochrane Centre has evaluated summaries of Cochrane reviews written for back pain sufferers [7] and the Cochrane Musculoskeletal Group has completed user testing with summaries for people with arthritis [8]. Based on that work and the work carried out by the German Cochrane Centre, it has become clear that the Cochrane Collaboration needs to develop better guidelines for authors writing plain language summaries. During 2008, with the support of the Cochrane Collaboration’s Opportunity Fund, six Cochrane entities will develop guidelines and formats for plain language summaries and test these among consumers and consumer representatives. This work will build upon a variety of experiences and research: the work carried out by the Norwegian Branch and the Musculoskeletal Group; guidelines from other organisations for writing patient information and research evidence for the presentation of information to patients and the public. Several issues for which there is evidence and other issues that have not yet been evaluated will be addressed during this work. For example, in current plain language summaries authors use a variety of words to express the strength of the evidence and the magnitude of the effect of the interventions. There is a need to develop standard qualitative phrases to avoid phrases such asy.’’a moderately large effect’’ or ‘‘a high likelihood of somewhat small but possibly important effects’’. Qualitative phrases do have shortcomings since they are interpreted differently by dif-
Z. Evid. Fortbild. Qual. Gesundh. wesen 102 (2008) 493–496 www.elsevier.de/zefq
ferent people [9]. But standard phrases, used by authors, consistently may decrease misinterpretation. In contrast, quantitative statements, that is, providing numbers to describe the evidence, give a more precise estimate of effect. Presenting numbers, however, has its own challenges. But despite the challenges, there are a few principles that can be incorporated into patient information. Natural frequencies are preferred (e.g. 5 out of 100). When expressing benefits or risks, it is better to not just describe the relative risks alone, but to include absolute risks, and to compare what happens to people with or without the intervention (e.g. 5 out of 100 had outcome without Drug A, 20 out of 100 had outcome with Drug A) [9]. There is also some evidence that people prefer to read information in tables [10]. Other issues, not yet evaluated will also be investigated when testing the various formats, such as how to present the quality of evidence and, if helpful, how to present confidence intervals, and how much information to give about the research that lies behind these results. Using these principles, a variety of potential formats for the plain language summaries have been developed. These formats will be tested in different countries and among different groups. This research should potentially be useful not only for the Cochrane Collaboration, but also for other groups preparing patient information. Once written, what more can Cochrane do? What more can other organisations do? The next step is to ensure that the information is used. Although a relatively new area of research, there is some evidence about how to promote the use of research information by patients. Cochrane and a number of Cochrane groups are already working on this knowledge translation to consumers.
How can we promote the use of plain language summaries? One of the simplest strategies, of course, is taking advantage of the
495
ARTICLE IN PRESS number of people accessing the Internet and seeking health information. As previously described, the Cochrane Collaboration has created a ‘safe site’ to search for and read the plain language summaries of Cochrane reviews. Many Cochrane Review groups also link to or have packaged these summaries on their own web sites for easy access (e.g. the Cochrane Wounds and the Cochrane Neonatal Groups) or produced separate patient information web sites (e.g. the Back Review Group www.rygginfo.no). Instead of creating a new site to house the plain language summaries, the Cochrane Musculoskeletal Group, for example, has taken knowledge translation one step further and provides its summaries on a variety of patient group web sites to directly reach patients and consumers. This group has been working with success with a couple of arthritis groups: The Arthritis Society in Canada (www. arthritis.ca/look at research/cochrane reviews and Arthritis Victoria in Melbourne (www.arthritisvic.org.au)). The summaries are stand-alone products on these web sites but are often incorporated into the general patient information provided on the site. Even as stand-alone products consumers visiting the Arthritis Society web site indicated that the summaries increased their knowledge and were useful for decision making [8]. Since nearly three quarters of people seek information from their physicians, targeting patients via their physicians is another strategy for knowledge translation. Evidence shows that patients benefit more from written information when used in consultation with health care professionals [11]. Research organisations who, as described previously, produce patient information alongside information for professionals, may increase the likelihood that physicians see patient information and use it in consultations. Decision aids, which present the information from the plain language summaries with tools to help
496
patients make health care decisions with their physician, have also proven to be an effective knowledge translation strategy. Decision aids have been shown to improve patient knowledge, and align the decisions that patients make with their values and expectations [12]. The Cochrane Musculoskeletal Group has produced and continues to produce decision aids based on their reviews and more research to evaluate their effectiveness is pending [8].
[2]
[3]
[4]
So, can we do it? Ultimately, the purpose of any translation of health research knowledge into formats that researchers, health care professionals, policy makers and patients can understand and use is to improve patient care. Informing patients directly makes sense, but it can be challenging. There is already some research evidence about how to make research information patientfriendly, but more needs to be done. There is also some evidence about strategies to ensure that the information is used. While it is important to build on and use this evidence, it is also valuable to learn from the experiences of others. The Cochrane Collaboration has over 15 years of experience incorporating the principles of how to present research to patients and facilitate its use. Therefore, we hope our experiences can feed into the existing research evidence, as well as into the efforts of other research organisations currently working in this field. In this way, we may create ‘evidence that patients can understand and use’. Funding: Cochrane Collaboration Opportunities Fund Canadian Institutes of Health Research
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
References [1] Ellins J, Coulter A. How engaged are people in their health care? Findings of a
national telephone survey. Oxford: Picker Institute Europe; 2005. Rutten LJ, Arora NK, Bakos AD, Aziz N, Rowland J. Information needs and sources of information among cancer patients: a systematic review of research (1980–2003). Patient Educ Couns 2005 Jun;57(3):250–61. N(o)nliner-Atlas 2007. TNS infratest, 2007. Available from /http://www.initiatived21. de/fileadmin/files/NOA_Umzug/NOA_ Atlanten/NONLINER-Atlas2007.pdfS. Eysenbach G, Powell J, Kuss O, Sa ER. Empirical studies assessing the quality of health information for consumers on the world wide web: a systematic review. JAMA 2002 May 22–29;287(20): 2691–700. Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.0.0 [updated February 2008]. The Cochrane Collaboration, 2008. Available from /http://www.cochranehandbook.orgS. Falck-Ytter C, Lang B, Booker D, Antes G, Falck-Ytter Y. The quality of Cochrane synopses – a quality assessment pilot project. 12th Cochrane Colloquium. Ottawa, 2004. Available from /http:// www.cochrane.org/colloquia/abstracts/ ottawa/P-040.htmS. Glenton C, Nilsen ES, Carlsen B. Lay perceptions of evidence-based information – a qualitative evaluation of a website for back pain sufferers. BMC Health Serv Res 2006 Mar 15;6:34. Tugwell PS, Santesso NA, O’Connor AM, Wilson AJ, Effective Consumer Investigative Group. Knowledge translation for effective consumers. Phys Ther 2007 Dec;87(12):1728–38. Lipkus IM. Numeric, verbal, and visual formats of conveying health risks: suggested best practices and future recommendations. Med Decis Making 2007 Sep–Oct;27(5):696–713. Schu¨nemann HJ, Sta( hl E, Austin P, Akl E, Armstrong D, Guyatt GH. A comparison of narrative and table formats for presenting hypothetical health states to patients with gastrointestinal or pulmonary disease. Med Decis Making 2004 Jan–Feb;24(1):53–60. Coulter A, Ellins J. Effectiveness of strategies for informing, educating, and involving patients. BMJ 2007 Jul 7;335(7609): 24–7. O’Connor AM, Stacey D, Entwistle V, Llewellyn-Thomas H, Rovner D, HolmesRovner M, et al. Decision aids for people facing health treatment or screening decisions. Cochrane Database Syst Rev 2003(2):CD001431.
Z. Evid. Fortbild. Qual. Gesundh. wesen 102 (2008) 493–496 www.elsevier.de/zefq