JMIR Publications

Select Journals for Content Updates

When finished, please click submit.

Citing this Article

Right click to copy or hit: ctrl+c (cmd+c on mac)

This paper is in the following e-collection/theme issue:

    Original Paper

    User-Generated Online Health Content: A Survey of Internet Users in the United Kingdom

    1Nuffield Department of Primary Care Health Sciences, University of Oxford, Oxford, United Kingdom

    2Medical School, University of Exeter, Exeter, United Kingdom

    3Department of Information and Communication Science, Universitat Oberta de Catalunya, Barcelona, Spain

    Corresponding Author:

    Braden O'Neill, BSc

    Nuffield Department of Primary Care Health Sciences

    University of Oxford

    Radcliffe Observatory Quarter

    Woodstock Road

    Oxford, OX2 6GG

    United Kingdom

    Phone: 44 (0)1865289414

    Fax:44 (0)1865289287

    Email:


    ABSTRACT

    Background: The production of health information has begun to shift from commercial organizations to health care users themselves. People increasingly go online to share their own health and illness experiences and to access information others have posted, but this behavior has not been investigated at a population level in the United Kingdom.

    Objective: This study aims to explore access and production of user-generated health content among UK Internet users and to investigate relationships between frequency of use and other variables.

    Methods: We undertook an online survey of 1000 UK Internet users. Descriptive and multivariate statistical analyses were used to interpret the data.

    Results: Nearly one-quarter of respondents (23.7%, 237/1000) reported accessing and sharing user-generated health content online, whereas more than 20% (22.2%, 222/1000) were unaware that it was possible to do this. Respondents could be divided into 3 groups based on frequency of use: rare users (78.7%, 612/778) who accessed and shared content less than weekly, users (13.9%, 108/778) who did so weekly, and superusers (7.5%, 58/778) who did so on a daily basis. Superusers were more likely to be male (P<.001) and to be employed (P<.001), but there were no differences between the groups with respect to educational level (P=.99) or health status (P=.63). They were more likely to use the Internet for varied purposes such as banking and shopping (P<.001).

    Conclusions: Although this study found reasonably widespread access of user-generated online health content, only a minority of respondents reported doing so frequently. As this type of content proliferates, superusers are likely to shape the health information that others access. Further research should assess the effect of user-generated online content on health outcomes and use of health services by Internet users.

    J Med Internet Res 2014;16(4):e118)

    doi:10.2196/jmir.3187

    KEYWORDS



    Introduction

    Background

    There can be little doubt that the Internet has changed the way that people experience health and illness. People routinely use the Web to learn about the meaning of symptoms, tests, diagnoses and treatments, and to find out how others have rated doctors and hospitals [1]. In the United Kingdom, more than two-thirds of residents use the Internet to obtain health information [2], and the UK Department of Health considers availability of high-quality information a key policy objective [3]. The Internet once provided information that could be accessed and viewed, but not easily modified. Patients increasingly go online to share information and to access information others have posted on the Internet, including their own experiences. Although this is a widespread practice, especially for long-term conditions [4], relatively little is known about the characteristics of people who share and access user-generated content (UGC) online. This is in contrast to general online health information-seeking behavior which has been extensively studied [1,3,4]. In this paper, we draw on a major European survey to explore who shares and accesses user-generated health content online in the United Kingdom.

    User-Generated Health Content

    User-generated health content is presented in many different ways across different types of websites [5]. Traditional, commercially produced online health content (developed by businesses or health systems or governments) has predominantly consisted of facts and figures, such as numerical data related to the diagnosis or prognosis of a condition. User-generated health information, by virtue of its distributed, often informal development and dissemination, is a broad concept and incorporates anything that a patient or carer posts online for the purpose of others seeing it. Entwistle et al [6] noted that online health information can be broadly divided into 2 categories: general facts and personal experiences. Although these categories are not mutually exclusive, the differences have important implications. General facts consist of information about particular conditions, treatments, or services, and are intended to be broadly applicable to many people interested in the same topic. Personal experiences may also provide information about conditions, treatments, or services, but are often in narrative form and portray an individual’s experience of undergoing a particular treatment or of living with a long-term condition. People tend to draw on both general facts and personal experiences, depending on the issue involved [6,7]. It has been noted that adequate information is an important aspect of health care delivery that matters to patients [8].

    General health information sites often provide UGC in the form of curated experiences selected by the site’s developers and the ability for site users to post their own comments. There are sites exclusively devoted to interview-based research on health experiences; for example, the UK’s Healthtalkonline has been available for more than a decade and now has many counterparts in other countries [9]. Blogs enable people to post their stories over time and provide the facility to embed video and links to other content, including peer-reviewed journal articles where applicable. Social networking sites enable patients to post content (both facts and experiences) and have pages devoted to specific conditions and treatments. Online forums host pages for specific conditions and facilitate patients coming together who have similar diseases or have undergone similar treatments. Some sites include crowdsourcing: “an approach to accomplishing a task by opening up its completion to broad sections of the public” [10]. For example, the website PatientsLikeMe aggregates data provided by site users. Patients can compare their data with aggregated input from other site users, and data obtained through crowdsourcing can be used for research [11]. Finally, there are those sites that enable a health service user to write and read reviews of specific hospitals, doctors, or health systems [12]. Increasingly, websites include several different ways of incorporating content from patients and the public. What all these disparate sources share is that their content depends on people posting, sharing, and comparing health information, whether about themselves or those for whom they care.

    Use and Nonuse

    One of the predominant places where UGC is accessed and shared online is in the context of virtual support groups for particular conditions as well as individual treatments. The heterogeneity of online support group membership has been reported in several publications; these groups consist of some people who post occasionally and some who are frequent prolific posters. Those who post have been described as key players [13], active users [14], and caretakers [15]. Most people who visit support groups do not actively post. In a study of people accessing an online smoking cessation group, 84.7% of those who registered and accessed the resource at least once never posted [16]. To explain this distribution, the “1% rule” has been adapted from marketing literature to the use of health social networks, suggesting that 90% of actors observe but do not participate, 9% rarely contribute, and 1% create most of the content [17]. This is contextual, and people’s behavior may vary between sites if they visit several different sites. Adams [18] has suggested that people may be more motivated to add their own commentary if they feel that their experiences are not covered by others accounts (eg, adverse drug effects). Because the content on these websites is generated by a small proportion of the people who use it, the concept of the “superuser” [17,19] of user-generated online health information—the minority of people who post information online—has received recent attention. This has not been characterized within the United Kingdom.

    The Aim of Our Research

    The purpose of this paper is twofold. First, we identify characteristics of UK Internet users who access and share UGC online. In this study, we use UGC to denote 4 specific behaviors about which questions were asked in the survey: (1) participating in an online support group for people who are concerned about the same health or medical issue, (2) participating in social networking sites talking about health and wellness, (3) describing a medical condition or problem online to get advice from other online users, and (4) disclosing medical information on social networking sites.

    Second, we investigate the frequency with which respondents engage in the aforementioned study behaviors, paying particular attention to those who do so most frequently because these users predominantly generate the content that others view. By grouping respondents according to how frequently they go online to access and share UGC, we aim to characterize the differences between these groups. We aim to determine areas for further research to support the effective and beneficial use of online health information.


    Methods

    Survey Instrument and Ethics

    We analyzed United Kingdom data from the Citizens and Information Communication Technology for Health survey, a project undertaken in 2011 by the Institute for Prospective Technology Studies of the European Commission’s Joint Research Centre. This online survey was developed from a theoretical framework of the social determinants of information and communication technology (ICT) for health, translated into native languages in 14 European Union (EU) member countries. The survey was developed to understand and characterize European citizens’ use of ICT for health. It included questions on a range of triggers, motivations, and behaviors. This analysis examines a subset of questions related to user-generated online health content. The full questionnaire is available in Multimedia Appendix 1 [20].

    Technical, methodological, and legal considerations were carefully addressed in the context of designing and implementing the survey. These considerations ensured anonymity and confidentiality of individual responses [21-23]. The survey was conducted in accordance with European Society for Opinion and Marketing Research ethical guidelines [24]. At the time the survey was carried out, one of the authors (FLV) was employed by the European Commission.

    Sample and Data Collection

    The sample consisted of 1000 respondents from the United Kingdom who completed an online survey in 2011 covering a variety of domains regarding the use of ICT for health. We randomly sampled people aged 16 to 74 years who had used the Internet in the 3 months before the survey to obtain the sample.

    An online invitation to participate in the survey was sent to 7291 individuals. The data were collected between July 20 and August 1, 2011. Simple random sampling was carried out by the software provider who administered the survey (Cint), then quota sampling was used to accurately reflect the demographic composition of UK residents [25]. We obtained 2410 responses (4881 nonresponses) of which 1320 were out of quota and 90 were excluded because of missing data. The average length of time to complete the questionnaire was 20.5 minutes.

    More than half (57.00%, 570/1000) of respondents were employed and nearly all (90.00%, 900/1000) had completed upper secondary or tertiary education (Table 1). This is broadly comparable to the characteristics of the UK population, although it should be noted that this sample is more highly educated than the general UK population [25].

    Table 1. Demographic composition of sample.
    View this table

    Statistical Analysis

    Initially, we calculated descriptive statistics about the characteristics of the study participants. After this, to determine the extent to which the 4 behaviors studied represent an underlying composite variable, we employed principal components analysis (PCA). In carrying out PCA, we a priori defined components explaining adequate amounts of variability in the data as those that have an eigenvalue greater than 1.00 [26]. Because the results of the PCA suggested that all 4 behaviors could be explained by one underlying component, for further analysis we treated this as a single composite variable, which we called “accessing and sharing user-generated health content online.”

    We then used nonhierarchical k-means cluster analysis to characterize respondents based on their reported frequency of accessing and sharing UGC online. We determined differences between the frequency groups by carrying out ANOVA and chi-square tests for categorical and continuous variables, respectively. Data were examined to determine if they met assumptions for the univariate and multivariate statistical tests carried out, and all assumptions were met. Statistical analyses were performed using SPSS version 20.0 (IBM Corp, Armonk, NY, USA).


    Results

    Frequency of Engaging in Individual Study Behaviors

    We found that approximately one-quarter of all respondents participated in online support groups (27.80%, 278/1000) and health-related social networking sites (23.80%, 238/1000) (Table 2). A similar proportion reported describing a medical condition or problem online to get help from other online users (23.10%, 231/1000) and disclosing medical information on social networking sites (16.70%, 167/1000).

    Less than 10% of respondents (4.80%, 48/1000 to 7.80%, 78/1000) reported engaging in any of the study behaviors at least or more than once a week. Approximately 20% of respondents (22.20%, 222/1000) reported being unaware that it was possible to engage in these behaviors online.

    Table 2. User-generated content behaviors.
    View this table

    Characterizing Predictors of Accessing and Sharing User-Generated Content

    Applying cluster analysis to the “accessing and sharing user-generated health content online” composite variable resulted in a 3-cluster solution. After excluding those who did not know that these resources were available (excluded because they have different reasons for nonuse than those who are aware of the possibility and choose to not engage) [26], we were left with 778 responses for further analysis.

    Most respondents (78.7%, 612/778) were rare users who reported that they never or very rarely accessed and shared user-generated health content. Users who went online and engaged in these behaviors infrequently (ie, weekly or monthly but not daily) accounted for more than 10% (13.9%, 108/778) of respondents. A further minority (7.5%, 58/778) were clustered as superusers who accessed and/or generated UGC daily.

    Sociodemographic Characteristics

    Comparing the 3 groups, superusers were more likely to be male than users or rare users (Table 3). Those in the 16-24 years age group were more likely to be superusers than users or rare users, whereas those in the 55-74 years age group were generally rare users, as opposed to superusers or users. Superusers were more likely to be employed (57.8%, 33/58) than unemployed (9.3%, 5/58), students (19.0%, 11/58), or retired/inactive in the labor force (24.0%, 13/58). The groups did not differ on level of education.

    Regarding household composition, respondents with children (1-parent, 2-parent, or extended family households) were less likely to be superusers than they were to be users or rare users. There were no differences between the groups based on the number of children younger than 16 years in the household. More rare users and superusers reported that they lived in intermediate areas, as opposed to densely or thinly populated areas, whereas users were more likely to report living in densely populated areas.

    Table 3. Sociodemographic characteristics of respondents (N=778).
    View this table

    Health Characteristics

    We did not identify any differences between groups for health status, undergoing long-term medical treatment, or number of visits to the doctor in the past 12 months (Table 4). Among superusers who reported having someone close to them currently experiencing long-term illness or disability, 64.3% (37/58) reported taking care of them.

    Table 4. Health characteristics of respondents (N=778).
    View this table

    Online Health Information and Health Behavior

    Superusers were more likely than users or rare users to report that the information they accessed affected the way they care for themselves and the way they eat or exercise (Table 5). They were also more likely to report that after accessing health information on the Internet, they subsequently talked to a doctor or nurse about what they found. Finally, superusers were more likely to report accessing information for their children than users or rare users.

    Table 5. Online health information and health behavior of respondents (N=778).
    View this table

    Internet Activity

    Almost all superusers (94.8%, 55/58) reported engaging in 11 or more Internet activities (eg, online banking, keeping a blog, and looking for travel information), whereas approximately one-quarter of rare users (26.6%, 163/612) reported the same engagement with other Internet activities (Table 6). Similarly, nearly three-quarters of superusers (74.1%, 43/58) reported accessing the Internet through 3 or more devices (eg, a home computer, work computer, and smartphone), which contrasts with less than one-third (29.4%, 180/612) of rare users doing the same.

    Table 6. Internet activities and devices of respondents (N=778).
    View this table

    Discussion

    Principal Findings

    Within the past few years, a rapid expansion of technologies that people can use to generate their own Internet content has provided novel opportunities for members of the general population to share and access health information. Social media have facilitated this ever-growing shift from the production of health information online being in the hands of commercial enterprise and health systems, to being led by users themselves. This study, the first to investigate user-generated health content on the Internet in the United Kingdom, sought to understand this growing trend through people’s responses to survey questions. One of the more striking findings in this study is that one-quarter of this sample of UK Internet users reported accessing and/or posting user-generated health information online. Because this is the first study of its kind in the United Kingdom, there are no equivalent data from which to study trends.

    As has been repeatedly shown in studies of online health support groups, most participants do not actively post their own content, although they read what others have written [13-17]. Although it is not possible to directly measure the activity of these “lurkers,” they represent such a large proportion of those engaged with UGC that we should see this as “normal” online behavior. Further exploration of the motivations and usage of UGC by these passively engaged individuals is required. By grouping responses to this survey according to the frequency with which people reported accessing and sharing UGC, we sought to understand the characteristics of people who are highly engaged in these behaviors—the superusers. Although our data do not provide information on the effects of UGC, our findings characterizing the 7.5% of respondents classified as superusers is valuable in describing the group who may be generating the content that others are viewing.

    These superusers are predominantly male. Respondents in the youngest age category (16-24 years) are more likely to be superusers, whereas those in the oldest age category (55-74 years) are generally rare users. Superusers carry out more varied activities on the Internet, such as online banking and booking travel, than the users and rare users. Despite these differences, our results suggest that there are no differences in health status or health service use between the superusers and the other 2 groups. This suggests that people who are well may be as responsible for producing our online health content as people who are ill. To our knowledge, this is the first explication that UGC does not appear to be associated with self-reported use of health services. This study did not distinguish between the use of information that provides guidance for a diagnosis or treatment of a health condition, and information shared or accessed by people who are healthy and either seeking or offering lifestyle advice. Several possible mechanisms have been identified by which UGC and, in particular, others’ experiences may affect health, such as finding information, feeling supported, and experiencing health services [5]. Therefore, our results are congruent with the contingent model of health information use, which suggests that health information is one important component of a health care experience, rather than as something that displaces or reduces use of health services [28].

    It was surprising that about 20% of participants were unaware of the availability of UGC on the Internet, particularly as this was a sample of Internet users. The outcomes of UGC are unclear; therefore, we would not necessarily recommend interventions to increase access. Further research should focus on reasons why these individuals, who have access to the Internet and are motivated to complete a health survey, are unaware of the availability of this increasingly ubiquitous content.

    Previous studies of health-seeking behavior on the Internet have found that females more commonly access health information on the Internet than males [1,4]. However, in the present study, superusers were more likely to be male. It has been suggested that men rate their self-assessed online skills higher than women do, which is one possible explanation for higher reported use of these complex online resources by men [29]. Although our data did not provide an opportunity to examine this issue further, it is an important area for future research as it guides content development and implementation.

    Limitations

    The survey used in this study was administered online. Administering a survey in this way enables rapid data collection, reduces administrative burden, and is cost-effective. It may cause selection bias because those who do not have Internet access as well as adequate digital skills are excluded [30]. In this study, rare users may have been excluded because of a lack of digital skills. Those with chronic illness may also have been less inclined to participate in an online survey or excluded because of fatigue or disability. Another limitation is the use of quota sampling (done to facilitate international comparisons in the overall survey), which reduces the generalizability of the results of this analysis to the overall population when compared with a true random sample. Although the sample was selected to be representative of UK Internet users, subsequent studies with different designs might sample purposively for superusers to explore their behavior and motivations more fully. Further research focusing on smaller age categories and going deeper into some of the life-cycle variables that may affect use of online health information (eg, parenting or retirement) would be useful as well. Furthermore, this study reports a secondary analysis of a larger survey on people’s use of the Internet for health. Some potential areas of inquiry relevant to UGC were not included in the original survey, such as its use to obtain social support and to select health services based on others’ experiences [5]. Finally, this study reports cross-sectional data, which precludes determining causal relationships. Because this is the first study of its kind in the United Kingdom, it is exploratory in nature and does not evaluate effects of UGC.

    Conclusions

    This study reports results of the first representative sample of UK Internet users that investigates accessing and sharing user-generated health content online. Within the context of available surveys from other countries, our results suggest that UGC may be increasingly popular among those who are healthy and who have chronic conditions alike, and that a minority of people who frequently access and share may be primarily responsible for generating the majority of content that others view.

    The potential benefits of ICT for health are vast, but it is likely that some online resources are effective and desirable for some people and they are not for others. Understanding the contexts in which they are helpful is important to be able to support individual patients, public health initiatives, and to develop information policy and strategy in clinics and health systems. Through understanding characteristics of those who already participate in accessing and sharing user-generated health content online, as well as the differences between groups of users based on the frequency with which they do so, the results of this study bring clarity to this important issue regarding use of ICT for health information. These findings inform an agenda for further research to identify why people access and share UGC, what the impact is on health behaviors and outcomes, and if expanding engagement with user-generated health content online is likely to be beneficial.

    Acknowledgments

    We would like to thank those who participated in the survey, as well as Ioannis Maghiros and Fabienne Abadie for their work on the original survey. We are grateful to Dr Brian Nicholson for his comments on earlier drafts of the paper. Braden O’Neill would like to thank the Rhodes Trust for their support of his studentship. This research was funded by The Institute for Prospective Technological Studies (IPTS), 1 of the 7 scientific institutes of the European Commission’s Joint Research Centre (JRC), and by the European Commission Directorate General for Communications Networks, Content and Technology (DG Connect) as a part of the Strategic Intelligence Monitor on Personal Health Systems II project (SIMPHS II). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. The views expressed are purely those of the authors and may not in any circumstances be regarded as stating an official position of the European Commission.

    Conflicts of Interest

    None declared.

    Multimedia Appendix 1

    "Citizens and Information Communication Technology for Health" survey.

    PDF File (Adobe PDF File), 6MB

    References

    1. Fox S. The social life of health information. Washington, DC: Pew Internet & American Life Project; 2011 May 12.   URL: http://www.pewinternet.org/Reports/2011/Social-Life-of-Health-Info.aspx [accessed 2013-12-10] [WebCite Cache]
    2. Dutton W, Blank G. Next Generation Users: The Internet in Britain. Oxford: Oxford Internet Institute; 2011.   URL: http:/​/oxis.​oii.ox.ac.uk/​sites/​oxis.oii.ox.ac.uk/​files/​content/​files/​publications/​oxis2011_report.​pdf [accessed 2013-12-10] [WebCite Cache]
    3. United Kingdom Department of Health. The Power of Information: Putting all of us in control of the health and care information we need. 2012.   URL: https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/213689/dh_134205.pdf [accessed 2013-12-10] [WebCite Cache]
    4. Fox S. Peer-to-peer healthcare. Washington, DC: Pew Internet & American Life Project; 2011 Feb 28.   URL: http://www.pewinternet.org/Reports/2011/P2PHealthcare.aspx [accessed 2013-12-10] [WebCite Cache]
    5. Ziebland S, Wyke S. Health and illness in a connected world: how might sharing experiences on the internet affect people's health? Milbank Q 2012 Jun;90(2):219-249 [FREE Full text] [CrossRef] [Medline]
    6. Entwistle VA, France EF, Wyke S, Jepson R, Hunt K, Ziebland S, et al. How information about other people's personal experiences can help with healthcare decision-making: a qualitative study. Patient Educ Couns 2011 Dec;85(3):e291-e298. [CrossRef] [Medline]
    7. Kamel Boulos MN, Wheeler S. The emerging Web 2.0 social software: an enabling suite of sociable technologies in health and health care education. Health Info Libr J 2007 Mar;24(1):2-23. [CrossRef] [Medline]
    8. Entwistle V, Firnigl D, Ryan M, Francis J, Kinghorn P. Which experiences of health care delivery matter to service users and why? A critical interpretive synthesis and conceptual map. J Health Serv Res Policy 2012 Apr;17(2):70-78 [FREE Full text] [CrossRef] [Medline]
    9. Herxheimer A, McPherson A, Miller R, Shepperd S, Yaphe J, Ziebland S. Database of patients' experiences (DIPEx): a multi-media approach to sharing experiences and information. Lancet 2000 Apr 29;355(9214):1540-1543. [CrossRef] [Medline]
    10. Ranard BL, Ha YP, Meisel ZF, Asch DA, Hill SS, Becker LB, et al. Crowdsourcing--harnessing the masses to advance health and medicine, a systematic review. J Gen Intern Med 2014 Jan;29(1):187-203. [CrossRef] [Medline]
    11. Frost J, Okun S, Vaughan T, Heywood J, Wicks P. Patient-reported outcomes as a source of evidence in off-label prescribing: analysis of data from PatientsLikeMe. J Med Internet Res 2011;13(1):e6 [FREE Full text] [CrossRef] [Medline]
    12. Adams E, Boulton M, Watson E. The information needs of partners and family members of cancer patients: a systematic literature review. Patient Educ Couns 2009 Nov;77(2):179-186. [CrossRef] [Medline]
    13. Cobb NK, Graham AL, Abrams DB. Social network structure of a large online community for smoking cessation. Am J Public Health 2010 Jul;100(7):1282-1289 [FREE Full text] [CrossRef] [Medline]
    14. Cunningham JA, van Mierlo T, Fournier R. An online support group for problem drinkers: AlcoholHelpCenter.net. Patient Educ Couns 2008 Feb;70(2):193-198. [CrossRef] [Medline]
    15. Jones R, Sharkey S, Smithson J, Ford T, Emmens T, Hewis E, et al. Using metrics to describe the participative stances of members within discussion forums. J Med Internet Res 2011;13(1):e3 [FREE Full text] [CrossRef] [Medline]
    16. Selby P, van Mierlo T, Voci SC, Parent D, Cunningham JA. Online social and professional support for smokers trying to quit: an exploration of first time posts from 2562 members. J Med Internet Res 2010;12(3):e34 [FREE Full text] [CrossRef] [Medline]
    17. van Mierlo T. The 1% rule in four digital health social networks: an observational study. J Med Internet Res 2014;16(2):e33 [FREE Full text] [CrossRef] [Medline]
    18. Adams S. Sourcing the crowd in healthcare and the hierarchy of pharmaceutical knowledge. Presented at: The Co-Production of Knowledge; July 18-20, 2012; York, UK.
    19. van Mierlo T, Voci S, Lee S, Fournier R, Selby P. Superusers in social networks for smoking cessation: analysis of demographic characteristics and posting behavior from the Canadian Cancer Society's smokers' helpline online and StopSmokingCenter.net. J Med Internet Res 2012;14(3):e66 [FREE Full text] [CrossRef] [Medline]
    20. Lupiáñez-Villanueva F, Maghiros I, Abadie F. Citizens and ICT for Health in 14 EU Countries: Results from an Online Panel. Seville, Spain: European Commission, Joint Research Centre, Institute for Prospective Technological Studies; 2012.   URL: http://ftp.jrc.es/EURdoc/JRC71142.pdf [accessed 2013-12-10] [WebCite Cache]
    21. Eysenbach G, Wyatt J. Using the Internet for surveys and health research. J Med Internet Res 2002 Nov;4(2):E13 [FREE Full text] [CrossRef] [Medline]
    22. Dillman D, Smyth JD, Christian LM. Internet, Mail, and Mixed-Mode Surveys: The Tailored Design Method. Hoboken, NJ: Wiley & Sons; 2009.
    23. Braithwaite D, Emery J, De Lusignan S, Sutton S. Using the Internet to conduct surveys of health professionals: a valid alternative? Fam Pract 2003 Oct;20(5):545-551 [FREE Full text] [Medline]
    24. Eurostat Statistical Office of the European Union. 2012. European social statistics   URL: http://www.esomar.org/knowledge-and-standards/codes-and-guidelines/guideline-for-online-research.php [accessed 2014-01-15] [WebCite Cache]
    25. European Social Statistics. Luxembourg: Eurostat Statistical Office of the European Union; 2012.   URL: http://epp.eurostat.ec.europa.eu/cache/ITY_OFFPUB/KS-31-12-666/EN/KS-31-12-666-EN.PDF [accessed 2013-12-10] [WebCite Cache]
    26. Jackson DA. Stopping rules in principal components analysis: a comparison of heuristical and statistical approaches. Ecology 1993:2204-2214.
    27. United Nations Educational, Scientific and Cultural Organization Institute for Statistics. 1997. International standard classification of education   URL: http://www.uis.unesco.org/Education/ISCEDMappings/Pages/default.aspx [accessed 2013-12-10] [WebCite Cache]
    28. Heaney D, Wyke S, Wilson P, Elton R, Rutledge P. Assessment of impact of information booklets on use of healthcare services: randomised controlled trial. BMJ 2001 May 19;322(7296):1218-1221 [FREE Full text] [Medline]
    29. Hargittai E, Shafer S. Differences in actual and perceived online skills: the role of gender. Soc Sci Quart 2006;87(2):432-448.
    30. Whitehead LC. Methodological and ethical issues in Internet-mediated research in the field of health: an integrated review of the literature. Soc Sci Med 2007 Aug;65(4):782-791. [CrossRef] [Medline]


    Abbreviations

    EU: European Union
    ICT: information and communication technology
    PCA: principal components analysis
    UGC: user-generated content


    Edited by G Eysenbach; submitted 13.12.13; peer-reviewed by T van Mierlo, M Broom; comments to author 12.01.14; revised version received 21.01.14; accepted 23.02.14; published 30.04.14

    ©Braden O'Neill, Sue Ziebland, Jose Valderas, Francisco Lupiáñez-Villanueva. Originally published in the Journal of Medical Internet Research (http://www.jmir.org), 30.04.2014.

    This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included.