Introduction 1 1 2 3 2 4 5 6 6 6 7 examine consumer attitudes to the availability and quality of Internet-based information on medicines; explore consumer reasons for using this information; explore consumer experiences in searching for and appraising information on medicines; investigate the self-reported impact and application of this information. This paper will present results from the broader study on consumer experiences in searching for and appraising Internet-based information on medicines. Methods Selection of Method 8 10 Research Instrument Table 1 11 13 The interview guide and questioning route was pretested with a convenience sample of consumers (N = 13) to test for interpretation, appropriateness, and comprehensiveness, and to establish face and content validity. No significant changes were made to the interview guide or questioning route as a consequence of this pretest. Participant Recruitment After approval was granted by the Human Research Ethics Committee of The University of Sydney, participants were enlisted for the focus groups by a recruitment agency. Participants were recruited from the agency's database of consumers across metropolitan Sydney, Australia via telephone using a screening questionnaire. 14 Study Design Six focus groups were conducted in a number of locations around metropolitan Sydney in March to May 2003. 11 15 11 10 The focus groups were facilitated by a skilled moderator while 2 assistant moderators observed and took notes. The group discussions were 1 to 1.5 hours in duration and were digitally sound recorded after permission was obtained from all participants. The recordings were transcribed verbatim. Participants also completed a demographics questionnaire that collected data on Internet usage. Data Analysis 16 17 Results Table 1 Demographics 18 Table 2 Table 3 19 19 Table 2 Participant usage of the Internet (N = 46 participants)     Characteristic Usage Frequency, Number of Participants Relative Frequency, % of Participants Length of experience in the use of the Internet                Location of Internet access                                              Searching for Internet-Based Information on Medicines Search Engines All participants had used a search engine to find information on medicines. Most participants had a single favorite search engine that they would always use, but a few reported using more than one search engine to find the information they required. The choice of search engines was determined by many different factors ranging from the default search engine on their browser to active selection based on self-developed criteria. Numerous participants were influenced by the search engine that was used by coworkers, for example: Group 4, Participant 8 I saw it on this guy's computer and . . . I thought 'Oh, I'm going to use this'. That's how I started it at work. Some participants also reported that their browser automatically defaulted to a certain search engine and a few participants were unable to identify the search engine they used, for example: Group 3, Participant 6 Couldn't tell you [the search engine] really. I just log on and use whatever comes on. Many participants used search engines recommended by family and friends. There were certain determinants that led some participants to actively choose a specific search engine. These included perceptions of the credibility of the search engine, ease of use, relation with services such as e-mail, and a lack of advertising. These determinants did not necessarily include perceived quality of the information on medicines obtained through their use. 20 Group 1, Participant 1 It's got an educational edge, that's my experience. When I was at university doing my second degree, that was one that was sort of promoted as credible I suppose. 21 22 23 Google was undoubtedly the search engine the majority of participants used most and preferred. This was especially true of the younger participants. The common perception was that Google appeared to be straightforward and did not focus on advertising, for example: Group 4, Participant 6 It's just got less [rubbish]. It seems to be direct to what you want. I think that other [search engines] always have these categories and they always have suggestions for buying things and stuff like that but Google's pretty much straight to the point. It's simple. Participants also commented that this search engine was useful for suggesting spelling corrections when errors were made, as medicine names were sometimes difficult to spell. A few participants reported preferring Google as their search engine of choice specifically for health-related searches but were unable to explain reasons for their preference. 24 25 26 27 28 Generally, although a variety of search engines were used by participants when seeking information on medicines, the majority of participants used the same few dominant search engines. Participants generally preferred search engines with less advertising, and would continue to use the same search engine if they were successful in their searches. Most participants used the same search engine that they used for nonhealth information, and were usually influenced by what was used by friends, family, and colleagues. Search Processes Participants displayed a large variation in the process of searching for information on medicines. Most participants found information by typing the name of the medicine (drug name or brand name) into the search engine. A few participants felt this was the only way of finding information on a medicine, for example: Interviewer How do you put in your searches? Group 6, Participant 2 Medicines are really specific to just the name. Other participants reported looking for broader information, for example: Group 1, Participant 2 I often use a more general [search]. I might use something like 'women's health' or something. And I like to see a whole range of things . . . rather than targeting specifically . . . and then I choose within that. The information found through this type of search was said to be less specific to one medicine and had more general or comparative information. Some participants used more-advanced search techniques such as quotation marks, phrases, and extra words to narrow down their searches. They displayed an understanding of how these techniques helped to focus their searches, for example: Group 4, Participant 6 If you type it in with quotation marks, it'll search for those words together whereas if you type them separately, it'll just search for them anywhere. Participants reporting advanced skills were generally observed to be those who were younger or those who had greater experience of the Internet through work or study. However, it was clear that search skills varied significantly. The following interchange illustrates the mixed levels of understanding as to how search engines work: Group 5, Participant 4 [You need to] ask a specific question . . . 'What are the side effects?' rather than typing in 'penicillin'. Participant 6 Yeah, you really have to do a whole sentence. A whole statement. Participant 3 I would type in 'penicillin side effects'. Participant 4 'Then it could hit on 'penicillin' or it could hit on 'side effects'. The uninformed way in which some participants agreed upon what they considered to be optimal search skills was obvious in the group discussions. The majority of participants in this study who reported searching using less-than-optimal techniques—such as typing in whole questions—tended to be nonworkers, for example, full-time homemakers or retirees. The search skills of participants varied widely and these differences may affect the resulting information that participants encounter. Searching via a search engine however, was not the only way of finding information on the Internet on medicines. Other Methods of Finding Internet-Based Information on Medicines Some participants mentioned ways of finding information on medicines in addition to using search engines. A few participants said that they guessed the Web sites of medicines by typing the name of the medicine in the address bar in the format of www.[brand name or drug name].com. Several participants found information on medicines from Web sites recommended by family and friends, and from seeing advertisements in seniors' and health publications. Some reported bookmarking favorite Web sites for future reference and a few subscribed to mailing lists at health-related Web sites. One participant described searching for information on medicines using online journals. Although aware that the information was not aimed at consumers, this participant still chose to use this means to search for pertinent information on medicines: Group 2, Participant 2 I actually searched via . . . the professional journals . . . And I guess that was a little bit harder to do it that way because . . . reading through the journals was quite difficult. I tend to just go to the abstracts. Participants reported using a variety of search skills to obtain information on medicines. However, the important issue was how they selected and appraised the information. Appraising Internet-Based Information on Medicines Selecting Internet-based Information on Medicines Participants described different ways of choosing which Web site to visit when selecting from the numerous results obtained from using a search engine. Some worked down the list of results from the first one while others looked for keywords in the Web site descriptions or for the Web site's recency. Often participants made a judgment based on the URL (Web page address) of the result, for example: Group 4, Participant 1 I actually like looking at the actual web address, just seeing how professional it is. Like if it's some silly thing, I won't bother going into it. Many participants also reported looking for indicators in the Web site address to determine whether it belonged to a government, a university, an official organization, or a pharmaceutical company. Even though most participants said they would not go beyond the first page of the search results, one expressed the opinion that the best information was in the middle of the results and not on the first few pages. This participant had the erroneous opinion that the first pages of results are older and that results appeared mainly in the order in which the information had been created. Many participants reported looking for the country of origin of the information and preferred information generated from their country of residence, for example: Group 4, Participant 2 If I'm searching for a medication . . . and it brought up some things and I noticed it was in Australia, I click on that. These participants felt that Australian information would be more applicable to them and professed an awareness of health-setting issues such as differences in the brand names and availability of medicines in different countries. However, others had more confidence in United States-based information because they believed that this was where most new research was undertaken. It was clear that most participants did not pay conscious attention to how they selected Internet-based information on medicines, with one referring to the process as "a vibe" that you obtain through experience. Another described this as a feeling that "things have a look of credibility." Similarly, many participants had trouble in articulating their selection process, for example: Group 5, Participant 3 I find that sometimes I get to a site and I think 'Gee, this is a good site, but I don't know how I got there.' You know what I mean? You fluke it. Despite the inability of many participants to express how they selected information on medicines, many were able to express what they would not select. Participants reported quickly rejecting sites that were slow to load, sites that contained too many graphics, and sites that had pop-up advertisements. The process of selecting information on medicines varied among the participants. It appeared that all participants had their own criteria for selecting and rejecting information which may or may not appear logical to others. Credibility of the source, however, appeared to be a common determinant in the criteria of all participants. Credibility of the Source of Internet-Based Information on Medicines Participants expressed conflicting opinions about the credibility of the source of Internet-based information on medicines. Many participants regarded information produced by pharmaceutical companies to be the "official" information on a medicine and therefore trusted this the most, while many others were suspicious of a possible information bias, for example: Group 1, Participant 7 If you're looking at [a pharmaceutical company website], they've got factories throughout the world, they're a pretty good company so . . . you know that they've done so much research it's credible information. Group 1, Participant 6 If it's a pharmaceutical company, they're gonna put a good stance on their drug. Many other participants preferred information that originated from what they considered to be impartial and reputable sources such as government, professional, or disease-focused organizations, or university Web sites. A few participants also reported looking for credentials such as the author's qualifications when assessing the credibility of the information provider. A small number of participants preferred information written by other consumers who had personal experiences in taking the medicine. However, most participants expressed that they would be less likely to trust information on medicines generated by other consumers, for example: Group 1, Participant 6 There are chat rooms . . . if you've ever been prescribed such and such a medication; you'll get people from all around the world . . . Participant 2 Do you not find that a bit dangerous because everything is rather specific to each person's body? Participant 6 Oh yeah, but it would be comparable to having a chat with some of your friends. Some participants felt that the authorship of Internet-based information on medicines should be regulated and feared the reliability of the information because there was "no watchdog" for the information published on the Internet while others regarded it as analogous to the way they would trust information given in common conversation and therefore felt comfortable using information in this context. The credibility of the source of information on medicines was a strong determinant in the selection process. However, in addition to the source participants evaluated information using criteria described in the next section. Evaluating Internet-Based Information on Medicines Participants evaluated information on medicines using criteria such as the motive for the information, the language used, and the applicability to their needs. Almost all participants were skeptical to some degree of Internet-based information on medicines. Many participants professed a universal need for consumers to inherently distrust this information, and to interpret it accordingly. One participant stated that it is important to also consider why the information is on the Internet: Group 1, Participant 6 What are the motives? Are they conflicting, credible? Whoever has posted it, are they trying to make a profit? Other participants described the obviously difficult-to-believe nature of some of this information and looked for signs of conspiratorial or misleading language when deciding whether to trust the information, for example: Group 2, Participant 8 If it says 'hazard free' and 'completely no side effects', for example, I'm more likely to disbelieve than believe that In addition to this awareness of unreliable information on medicines, many participants also expressed an understanding that the information they find may not necessarily be applicable to them and that the information should not be used at face value, for example: Group 2, Participant 7 The thing with medicines is there's no sort of right or wrong . . . Everyone's different, everyone's going to have a different reaction. Group 2, Participant 6 When you ask the doctor, they tell you 'well, [the side effects] happen but it's not like that', I think what happens is that the information is not tailored for myself. It's general information. Pertinent to this appraisal was the information-filtering process described by participants: Group 2, Participant 4 It's always better to try and take as much information and try and sift out what's useless Group 1, Participant 6 When they're talking about people using this medicine, 'ninety-eight percent will die within five years' . . . you have to take that and filter it through a whole bunch of other variables . . . and whether [the information] is not terribly well informed or completely informed. One common way in which some participants were able to filter information on medicines was to use other Web sites for comparison and cross-checking, for example: Group 4, Participant 1 I always go to two or three sites. Although participants reported methods of evaluating information, many expressed a difficulty in their evaluation, for example: Group 2, Participant 4 How do you [figure] out what's useful? Group 3, Participant 7 How do you know what's reliable and what's not? Ultimately, despite an awareness of the shortcomings and difficulties in evaluating the quality of information on medicines, all participants saw the Internet as an important resource for this information, for example: Group 1, Participant 1 I think as patients you expect immediate information and the Internet, whether it's credible or not, it's the fact that people can get it. Discussion 7 29 32 33 29 34 36 29 31 37 40 7 6 6 41 42 43 44 47 Limitations in This Study There are several important limitations in this study. 6 Second, the bounded period of 12 months in the inclusion criteria may be too long for consumers to correctly remember details of how they searched for and chose information. It might have been beneficial to actually perform a search as an activity to stimulate the participants' memories. Third, participants in group situations may feel compelled to provide socially-desirable answers that are not necessarily accurate. In this study, we sought to minimize this by informing participants that their results would be confidential and that they were welcome to speak about anything they felt even if they disagreed with someone else. However, this does not negate the problem. Although the use of individual interviews may help to minimize this discrepancy, this method is more time-consuming and cannot use group interaction for the generation of ideas. Last, certain actions are intuitive and therefore difficult to articulate. Most participants were not able to adequately describe their search and appraisal processes, which suggests that this process may largely be a form of tacit or implied knowledge. Therefore, future research needs to take into account actual observed (rather than reported) search and appraisal skills of consumers who are seeking information on medicines for their own use. Conclusion and Future Research 48 49 19