A content analysis of direct-to-consumer (DTC) prescription drug web sites.

While direct-to-consumer (DTC) prescription drug advertising in the United States continues to grow exponentially, all stakeholders continue to either question its value or fight for its continued existence. Legislators, drug manufacturers, advertising agencies, the Food and Drug Administration

(FDA), managed care organizations, and consumers are currently debating the value of advertising prescription drugs directly to consumers (AMA 1997; Cohen 1988; Friedler 1997; FTC 1996; Law 1998; McCarthy 1998; Nader 1999; Peterson 1998). Perhaps no other form of consumer advertising has caused such heated debate in recent history. Regardless of one's viewpoint, more research is needed to answer important questions surrounding this form of communication. The FDA "encourages sponsors and other interested parties to make known their research relating to the overall effects of DTC promotion on public health" (FDA 1999, p. 4). A better understanding of the content of DTC Web sites is likely to prove essential to creating public policy in this area because of its ability to communicate the vast amounts of information necessary to meet FDA guidelines and fully inform the consumer.

Interest in DTC drug advertising has increased almost as rapidly as the sales metrics. In 1991, advertisers spent $55 million on DTC advertising (Prescription Access Litigation 2002), and by 2000, annual spending on such advertising rose to $2.5 billion (Kaiser Family Foundation 2001). The number of prescriptions dispensed per person annually rose from 8.7 in 1996 to 10.8 in 2000 (Fetto 2002). A consumer survey found that 30% of respondents have talked to their doctor about a specific drug they saw advertised, and of those, 44% say their doctor gave them a prescription for that drug (Fetto 2002). A national FDA survey yielded slightly different results--25% of doctor visits included a specific drug request and 69% of those received a prescription (Kelley 2002).

Recent empirical findings support the need for research on health issues and the Web because of its increasingly pervasive impact. A 2003 survey found 80% of all patients go online for health-related information (Von Knoop et al. 2003). Of those who go on-line, 90% said the information they found on the Web improved their understanding of their health problems. Finally, one segment of health consumers (who desire the greatest control in their health care) prefers to get their health information from the Web (46%) rather than from their doctor (45%) (Von Knoop et al. 2003).

In general, past studies regarding DTC prescription drug advertising focused on the following areas (references available upon request: attitude toward DTC prescription drug advertising, consumer recognition and awareness of DTC prescription drug advertising, effects of DTC prescription drug advertising on consumers, physician attitudes and behaviors concerning pharmaceutical drug advertising, public policy implications of DTC prescription drug advertising, overall effects of DTC on one or many stakeholders, message analysis and a beginning look at how people use DTC Web sites (Maddox 1999). However, the content of DTC Web sites has received little attention and represents a key component that DTC marketers and academicians need to understand, because the FDA requires that the pharmaceutical company make complete information available (FDA 1999) and the company's Web site is the most common medium for doing so. This research is an important exploratory step in understanding what is being communicated to consumers on these Web sites, how the Web fits into the DTC communication structure, how DTC Web sites may or may not be adequately serving the FDA's policy, and what policy changes may be recommended. To aid this evaluation, it is necessary to analyze the content and format of DTC Web sites.

Therefore, the central research question of this study is, What is the form and content of DTC drug Web sites? The broad areas of analysis include the use of inducements, advertising appeals, inclusion of medical/drug information, and message format. The importance of this research is in its application to public policy issues of DTC advertising. The Web is an essential component of the DTC drug companies' communication with consumers, as many television and print advertisements refer consumers to the Web for additional information and use this to fulfill the FDA's requirements. Furthermore, the policy implications of this research extend beyond the United States. Recently, the European Union (EU) initiated a limited experiment of DTC advertising by providing information to consumers on treatments for three therapeutic categories (diabetes, AIDS, and asthma) through pamphlets and Web sites only (Calfee 2002). This clearly indicates the global importance of understanding DTC Web sites in creating public policy.

This study is a census of DTC Web sites and thus provides a thorough description of this type of DTC communication. Another strength of this study is that it adds to a previous medical information typology (Bell, Wilkes, and Kravitz 2000) and provides a complete list of medical information content, similar to that of Resnik and Stern's (1977) classification of television advertising information content. This would help the growing international "interest in the development of quality consumer health information" (Coney 2002, p. 220).

LITERATURE REVIEW

The majority of the literature on DTC drug advertising has focused on the value it may or may not serve in the health care process. Thus far, the vast majority of the research has investigated print and, to a smaller extent, television. To form a more complete picture of the value of DTC drug advertising, it is important to better understand what role the Web plays in communicating DTC drugs to the consumer. Therefore, the research objective is to understand how these sites are communicating with the consumer (inducements, medical and drug information, advertising appeals, message format, and use of interactivity and multimedia). Thus, this research provides understanding of how the Web may serve to meet FDA guidelines and whether it adds value to DTC campaigns. The key literature regarding DTC drug advertising will be reviewed below and includes the role of the FDA, the content of print advertisements, and the potential role of the Web.

Brief History of DTC Drug Advertising and the FDA

Traditionally, pharmaceutical manufacturers directed their prescription drug advertising toward physicians. This makes sense because physicians have the final authority to prescribe drugs in the United States. Companies began to advertise directly to consumers in 1985 (Alperstein and Peyrot 1993) when the FDA lifted the ban on DTC advertisements. DTC advertising began to appear at that time, but it was minimal. The FDA had no specific regulations for DTC advertising, so the ads were subject to the same regulation as ads for health care professionals (People's Medical Society 1996), including the technically worded "brief summary" (i.e., verbose prescribing information regarding the prescription drug).

In August 1997, the FDA lifted the requirement to include the "brief summary" in broadcast advertising, and immediately thereafter drug manufacturers began to expand their television campaigns (FDA 1997). In addition, the FDA made it clear that advertisers may omit the brief summary if they make "adequate provision ... for dissemination of the approved or permitted package labeling in connection with the broadcast presentation" (p. 3).

In 1999, the FDA outlined the approach "that the FDA believes can fulfill the requirement for adequate provision in connection with consumer-directed broadcast advertisements for prescription drug and biological products" (p. 2). The FDA presumes that the ads

1. Are not false or misleading in any respect.

2. Present a fair balance between information about effectiveness and information about risk.

3. Include a thorough major statement conveying all of the product's most important risk information in consumer-friendly language.

4. Communicate all information relevant to the product's indication (including limitation to use) in consumer-friendly language. (FDA 1999, p. 2)

The FDA's requirement of fulfilling the "adequate provision" entails an approach that "will allow most of a potentially diverse audience to have reasonably convenient access to the advertised product's approved labeling" (FDA 1999, p. 2). The FDA then goes on to explain different ways the provision may be met: toll-free telephone numbers, print advertisements appearing concurrently with a broadcast campaign with the prescribing information available, disclosure that pharmacists, physicians (or other health care providers), or veterinarians may provide additional product information, and, finally, the disclosure of an Internet Web page (URL) address that contains the additional information (FDA 1999). Over the course of these policy changes, the increasing emergence of DTC drug advertising caught the attention of several researchers who began to examine its content and policy implications.

Content of DTC Advertising

In one of the first studies, Roth (1996) identified several patterns in DTC print advertising through the use of a content analysis. Pharmacists served as judges in this study and reported the majority (two-thirds) of the advertisements as having "fair balance" of benefit and risk information and that the advertisements communicated in a way that consumers are likely to understand. However, most of the advertisements omitted information important for proper use by the consumer and Roth (1996) suggested that the FDA consider including additional information on misuse. Two additional sources suggest how to evaluate the quality of consumer health information. Although both of these were created in the United Kingdom and were originally designed to be applied to general health information instead of DTC advertisements, they are useful guidelines to evaluate DTC advertising. Coulter, Entwhistle, and Gilbert (1998) included the following in their guidelines (see Coney 2002 for a complete list):

1. Ensure that common concerns and misconceptions are addressed.

2. Refer to all relevant treatment or management options.

3. Include honest information about benefits and risks.

4. Include checklists and questions to ask the doctor.

5. Include sources of other information.

Specifically designed to evaluate information on the Internet, DISCERN developed questions to evaluate the quality of information on treatment choices, including the following (see Coney 2002 for a complete list):

1. Does it describe how each treatment works?

2. Does it describe the benefit of each treatment?

3. Does it describe the risks of each treatment?

4. Is it clear that there may be more than one possible treatment choice?

Coney (2002) states that in New Zealand, "DTC advertising performs badly at presenting evidence-based information about risks and benefits" (p. 221), and much of the U.S. research/ opinion would concur. However, the Web's ability to include more depth and breadth of information may be key to communicating this vital and detailed information to consumers.

Parker and Delene (1998) further examined print DTC advertisements to better understand the marketing and public policy implications. They found that the use of FDA-mandated product safety disclosures by advertisers increased from 63% in 1992 to 96% in 1995 (Parker and Delene 1998). Parker and Delene published another study in 1999 that focused on the types of messages DTC prescription drug advertisements were communicating from their headlines and subheadlines (Parker and Delene 1999), which included the following: news/feature, problem/solution, testimonial, endorsement/authority, education, and humor. The authors found that problem/solution was the most popular appeal type and humor was the least popular (Parker and Delene 1999).

Research knowledge about DTC print advertising was furthered in several ways by Bell, Kravitz, and Wilkes (2000). First, they investigated inducements (e.g., coupons, rebates, etc.) to see how frequently and what type of incentives were offered to the consumer to attempt to get them to act (e.g., try the drug, talk to their doctor, get more information). They felt inducements were particularly important because they might coerce the patient to choose that drug without fully evaluating the alternatives. They found that 17% of the DTC print ads included inducements (Bell, Kravitz, and Wilkes 2000). Second, they developed and utilized a taxonomy of advertising appeals or selling points that contained four main groups: (1) effectiveness, (2) social-psychological enhancements, (3) ease of use, and (4) safety (see Bell, Kravitz, and Wilkes 2000 for complete definitions). They found the most common appeals included "effective" (57%), "controls symptoms" (41%), "innovative" (41%), and "convenience" (38%). Appeals found less commonly included "nonmedicated" (14%), "safe" (11%), "quick acting" (6%), "lifestyle" (6%), and "easy on the system" (3%). In addition, certain illnesses were more likely to use particular appeals (e.g., allergy with symptom control, cancer and cardiovascular disease with prolonged life, HIV/AIDS with powerful treatment, etc.).

Understanding the use of advertising appeals is important when evaluating DTC on the Web, because it indicates whether the priority is selling or educating. Education of consumers about the medical condition and specific drug was the focus of a separate study by the same researchers (Bell, Wilkes, and Kravitz 2000), in which they developed a classification of medical and drug information and identified their prevalence in DTC print ads, including condition name (95%), misconceptions (9%), precursors (27%), prevalence (12%), symptoms (60%), alternative treatments (29%), mechanism of action (36%), success rate (9%), supportive behaviors (24%), time to onset of action (20%), and treatment duration (11%). This 14-category classification will be applied to the present study.

A more recent content analysis further investigated how well DTC prescription drug print advertisements were informing the consumer, as well as what appeals were used (Woloshin et al. 2001). Emotional appeals were found 67% of the time and experiences perhaps caused by medical reasons were found 39% of the time. Vague and qualitative language to describe the benefit of the medication was found in 87% of the advertisements. Data used to describe the benefits of the drug were only used in 13% of the advertisements. About 50% of the advertisements did use data to describe side effects and listed side effects that generally occurred infrequently. The authors concluded that "provision of complete information about benefit would serve the interests of physicians and the public" (Woloshin et al. 2001, p. 1146).

To further what is known about DTC advertising, it is important to apply many of the concepts reviewed here to DTC on the Web to form a more complete picture and help guide public policy. This leads to the first research questions.

RQ1: How are DTC prescription drug Web sites communicating with the consumer?

a. To what extent do they use inducements and what sort of inducements do they use?

b. What medical and drug information is included in the sites?

The Importance of DTC Communication on the Web

A common argument used by stakeholders interested in continuing to advertise prescription drugs directly to consumers is that this type of communication fosters education regarding medical conditions and their treatments (Wilkes, Bell, and Kravitz 2000) and that consumers need to be better informed and take a more active role in their own health care (Calfee 2002). For many, the Web is the main source of health information that they can actively seek out to become better informed and educated. Although some studies have indicated that print advertisements do not fully educate consumers (Wilkes, Bell, and Kravitz 2000), many times, secondary sources of information are disseminated through the advertised communication. For example, 14% of DTC advertisements provided a Web site address in 1996, but by 1998, 57% did (Wilkes, Bell, and Kravitz 2000). Wilkes, Bell, and Kravitz (2000) recommended that future research evaluate the secondary sources of information available to consumers, including the Web, to better understand their potential to facilitate education.

Compared with the automotive and consumer electronics industries, the pharmaceutical industry is only spending a fraction of the total amount spent on promoting via the World Wide Web, but it is becoming increasingly important. As reported by a pharmaceutical advertising publication, the Web was not even listed as part of the $2.24 billion spent in 2000 on DTC advertising, which listed outdoor last with just over $2 million (Scussa 2001). In 2001, an advertising industry publication reported the following breakdown: television had 67.4% of DTC ads, followed by magazines (42.7%), radio (29.8%), newspapers (25.7%), Internet (17.6%), Sunday newspaper magazines (16.6%), and outdoor (9.5%) (Neff 2002). The differences in allotment to the Web are possibly due to both the increase in the Web's importance and differing views of its role as a DTC communication (i.e., advertising or not) by the two different industries.

Research is also indicating that consumers are using the Web to do research about their medical conditions and treatment options. It is estimated that by 2005, more than 80 million Americans will do such searches (Twibell and Anzalone 2001). It is also less expensive for pharmaceutical companies to promote via the Web than via traditional media. In addition, about 25% of on-line information is related to health (Matthews 2001). All this indicates the increasing importance of the Web to DTC advertisers and health communicators. Furthermore, Green (1998) suggested that the Internet is better suited for health care products than other media because of its ability to comply with regulatory disclosures.

Advertising Appeals

It is important to extend what is known about print DTC advertising appeals to the Web to better understand whether pharmaceutical companies are utilizing the Web as an educational tool or as a persuasive medium. Although the appeals used in DTC advertising are somewhat unique to the product category, it is helpful to place them in the larger context of advertising appeals that has received extensive research attention. Advertising appeals tend to be either rational or emotional and are aimed at persuading the individual to take some action (e.g., purchase, obtain more information) or make a cognitive change (e.g., awareness, attitude change). Emotional appeals include warmth, humor, and fear. Rational appeals include comparative advertising, open versus close-ended comparisons and one versus two-sided comparisons. In a more general way, appeals tap into our human nature and desires for sustenance, security, rest, love, social worth, independence, success, power, convenience, curiosity, fear of loss, health, kindness, and mental stimulation (Burton 1990; Russell and Lane 2002). As detailed above, the studies that specifically look at advertising appeals used in DTC drug ads (Bell, Kravitz, and Wilkes 2000; Woloshin et al. 2001) reported drug appeals most often associated with emotion (particularly fear, convenience, safety, and social benefits). Regardless of the type of appeal used, its presence indicates the desire to persuade in some way. Therefore, the following research question is posed.

RQ1.c: What advertising appeals do DTC Web sites use and to what extent do they use them?

Message Characteristics

Another area that has a long history of advertising research but has received little research attention in the area of Web DTC drug advertising is message characteristics, such as format, picture content, and interactivity. These are essential components to understanding how DTC Web sites are communicating and persuading consumers. This point was recently made in an industry report that stressed the importance of using interactive tools for e-health sites to be competitive and successful (Von Knoop et al. 2003).

Research on print ads has focused on the importance of copy and pictures. Some key findings from these studies include the recall benefit of paring pictures and words, the importance of consistency between copy and pictures, and the persuasive impact of pictures. The message characteristics of DTC ads have only recently been examined. For example, a study recently found the format of risk information (using color to distinguish from other text) facilitated knowledge acquisition (Wogalter et al. 2002).

Interactivity is a message characteristic that has received increased attention due to the Web. McMillan and Hwang (2002) recently summarized the definitions of interactivity and developed a measure of perceived interactivity based on three main features: direction of communication, user control, and time. The potential positive and negative effects of interactivity (e.g., attitudes, telepresence, decision making, comprehension, memory, purchase intention, and customer experience) make it an important element in studying DTC Web sites and ascertaining public policy implications (Coyle and Thorson 2001; Haubl and Trifts 2000; Li and Bukovac 1999; Macias 2003; Novak, Hoffman, and Yung 2000). The following research questions will address these issues.

RQ1.d: What is the message format in terms of graphics, types of pictures, and so forth

RQ1.e: Are the message formats utilizing the interactive and multimedia capabilities of the Web?

RESEARCH METHOD

This content analysis study utilized a census and made every effort to include all identifiable, working DTC prescription drug Web sites as of March 2001. An exhaustive search was conducted to identify all stand-alone Web sites for DTC drugs (e.g., www.claritin.com) (i.e., sites that were for the drug itself as opposed to part of the manufacturer's site) by seeking out on-line (e.g., Webmd.com) and off-line (television) ads for DTC drugs, searching the pharmaceutical companies corporate Web sites for links to stand-alone DTC drug sites, using trade and journal articles listing sites (Bell, Kravitz, and Wilkes 2000; Humphreys 2000; Realto 1998b), and searching on-line resources (e.g., www.rxlist.com/top200 .htm, www.pickdrugs.com/manuf_brands.htm, health .excite.com). After editing the list to remove any over-the-counter drugs or other sites that did not meet the criteria of this sample (e.g., manufacturer homepage), the population from which the data in this study were collected included 97 DTC drug Web sites.

Sample

This study employed a census of complete DTC drug Web sites, which was a total of 90. Seven sites had to be removed from the sample because they were not DTC drug sites, were not working, or needed a password. This indicates that DTC drug advertisers are utilizing the Web to a high degree. Two sources have reported the number of DTC brands as being close to 100. An academic study of DTC drug print ads (Bell, Kravitz, and Wilkes 2000) found 101 different brands. In addition, an industry source, MedAdNews, reported there are 103 DTC drug brands (Scussa 2001). Although this census may have missed some brands, and surely some drugs have come and gone, it seems clear that comparing this study's sample of 90 working Web sites is impressively close to 101/103.

The medical conditions spanned the entire range, with OB/GYN and psychiatric/neurological being the most common (see Table 1 for complete breakdown). The "other" category proved to be very important in this case, as several of the sites did not fit into the previous taxonomy (e.g., pain, weight loss, hair loss, and motion sickness). The drugs in the sample are manufactured by 28 different companies, with the most common being Ortho-McNeil Pharmaceuticals and AstraZeneca (eight each), followed by Glaxo Smith Kline, Pfizer, Pharmacia, and Roche Labs (seven each).

Code Sheet Development

A code sheet was developed to record information on how and what DTC drug Web sites were communicating with the consumer. The extensive code sheet included variables used in previous content analyses of DTC drug print advertising (Bell, Kravitz, and Wilkes 2000; Bell, Wilkes, and Kravitz 2000; Wilkes, Bell, and Kravitz 2000), as well as variables appropriate to the new medium (e.g., interactivity and multimedia). Specific coding topics included brand name, URL, manufacturer, medical condition, inducements, advertising appeal/selling points, medical information, and message characteristics. The medical information presented in the Web sites was coded according to the method used by Bell, Wilkes, and Kravitz (2000), as discussed in the literature review, with a few additions (prescribing information, side effects, and contraindications discussed in Web site content).

The next section of the code sheet captured the message characteristics, including people/animals pictured (doctor, pharmacist, man, woman, child, animal, and other), format of the message (copy, pictures, graphics, color, animated graphics, video, audio and language option), interactive features (e.g., registration, submit a question, hyperlinks, search features, games, etc.), and any additional information included in the site (e.g., copies of ads, statement of privacy, FAQs, site map, etc.). An "other" category was included for interactive features because the possibilities could be vast and somewhat specific to the drug being advertised. After the coders were trained (described below), they pretested the code sheet to establish preliminary reliabilities and determine whether modifications needed to be made to the code sheet or definitions.

Procedure

Two coders (one undergraduate and one graduate advertising major at a southeastern state university) were thoroughly and extensively trained on how to code the Web sites and what the definitions were for each variable on the code sheet. All Web sites in the sample were saved as a complete site on zip disks for each coder so that they were both accessing the same information. This was a crucial step, as the content on the Web is in constant flux (Weare and Lin 2000). The coders were instructed to access the sites from their zip disk and code all sites from the same type of computer (i.e., high speed connection, 17-inch monitor) so that the Web site content they viewed would be as similar as possible (as recommended by Weare and Lin 2000). The unit of analysis in this study was a complete DTC drug Web site, defined as all HTML pages that are included in the site's root (e.g., www.viagra.com). Therefore, the entire site was coded, including all links and pages associated with that particular site without going beyond the main root address. This method has been used in previous studies to obtain the most comprehensive snapshot of the site (Chan-Olmsted and Suk Park 2000). The coders were instructed to explore the site thoroughly and make sure that all portions of the site were viewed by using the site map as a final check. McMillan (2000) found that many Web studies only code the front page of the site; however, the authors believed coding the entire site was crucial to getting the clearest picture of how the Web is being used by pharmaceutical companies to communicate to consumers. The Web sites were saved on June 20, 2001, and the coding was completed from June through August 2001.

Reliability

Two trained coders independently coded the Web sites. Intercoder reliability was established using the coefficient of reliability (Holsti 1969; Krippendorff 1980; Rifle, Lacy, and Fico 1998). Web sites were independently coded with 50% of the census sites being coded by both coders to establish reliability (each coder then coded the remaining one-quarter of the sites that the other did not code). The overall coefficient of reliability was 84.8%, and ranged from 100% (for variables such as "inclusion of audio" or "safety, nonmedicated" selling point) to 50% (for "inclusion of graphics"). Other variables that had the greatest amount of disagreements included "effectiveness, symptom control" (63.2%), "symptoms of the condition" (68.4%), and "link to the manufacturer" (68.4%). The overall average is above the minimal 80% agreement level (Riffe, Lacy, and Fico 1998). This is considered to be quite good for this study because it included interval-level data (e.g., ranges to code the number of graphics, pictures, etc.). All variables with agreement below 70% were dropped from the analysis. The two coders resolved all disagreements and 100% agreement was achieved.

RESULTS

Through the literature review, several areas were identified as being key to understanding how and what DTC Web sites are communicating to consumers and the corresponding public policy implications. This section profiles how DTC drug Web sites are communicating with consumers (RQ1) and the Discussion section will extend the public policy implications by providing understanding of the role of the Web in DTC advertising.

To What Extent Do DTC Drug Web Sites Use Inducements and What Sort of Inducements Do They Use?

As mentioned earlier, inducements are particularly important to public policy because they may coerce the consumer to act based solely on the inducement as opposed to a complete evaluation of all available information. Inducements most often involve monetary rewards. Rebates were identified in 14% of the Web sites, coupons in 8%, and free trials in 7%. Money-back guarantees were not present in any of the Web sites. However, the "other" category for monetary incentives captured many different ideas designed to give consumers something of value and encourage them to use the drug and/or establish a relationship with the company. Twenty-three percent of the sites included such offers--"money-saving offers," binder, monograph, book, free 60-minute phone card, "buy four get one free" diary, exercise kit, reimbursement, free customized homepage to chart cholesterol, free hands-free cell phone adapter, free tickets, free travel guide, medicine case/key chain, nasal allergy kit, $100 gift certificate to do a survey, awards, scholarships, offers/events, free magazine subscription, health/beauty samples, and sweepstakes entries. Therefore, in total, 52% of the sites included some type of monetary incentive.

Offers to send or provide additional information can also be important inducements. Only 4% offered to send a video, none of the sites offered to send audiotapes, 32% offered to send print materials, and 11% offered to send "other" information (primarily via e-mail). Combining these, 47% of these sites offered to send additional information.

Finally, the least common type of inducement offered was patient support programs. Only three sites offered these services, including "kids unlimited patient program," support network and weight management program, and the Zyban advantage plan (customized e-mail support system). However, in many respects, a large number of these Web sites are designed to provide support services by providing a place to ask questions and obtain information. Overall, inducements are clearly an important component of these sites, particularly monetary incentives. This indicates that a goal of these sites is to encourage the consumer to act by getting more information, talking to his or her doctor, or requesting a prescription for the drug.

What Medical and Drug Information Is Included in the Sites?

Although it is beneficial to understand to what degree inducements are being utilized in these sites, the medical and drug information presented on them is clearly a topic of great importance given the health implication of taking or not taking the drug. The educational value of these sites becomes increasingly apparent when the medical information, which is a large portion of the informative material, is observed. As Table 2 shows, a high proportion of the DTC sites are including key pieces of medical and drug information; the highest include condition name, mechanism of action, prescribing information, and side effects. The only two pieces of information that are included by less than 50% of the sites are condition misconceptions and success rate.

Overall, the content of DTC sites is aimed at informing the consumer about the medical condition and the brand being advertised. The vast majority (94%) of the sites encouraged the patient to talk to his or her doctor or pharmacist. The results also indicate that these sites are meeting several suggested guidelines of high-quality health information (Coulter, Entwhistle, and Gilbert 1998; DISCERN, as listed in Coney 2002), including having a form to print out that would help patients talk to their doctor and ask questions (43% of sites), discussing misconceptions (21%), describing how the treatment works (83%), and discussing alternative treatments (63%). Although content is of utmost importance for DTC drug Web sites, the advertising appeals are also a main concern for advertisers.

What Advertising Appeals Do the Sites Use?

As seen in Table 3, the most commonly used advertising appeals or selling points were "effective" (63%), "convenience" (52%), "lifestyle enhancement" (44%), "innovative" (27%), "safe" (24%), and "prevention" (22%). Overall, the advertising appeals and selling points focus on the aspects of the drugs that are most beneficial to the consumer, such as effectiveness, convenience, lifestyle enhancement, and innovativeness.

What Is the Message Format of the Sites, and Do They Use Multimedia Elements?

As was previously mentioned, the delivery of the message was categorized into four main areas: humans/animals pictured, format of the message, interactive features, and additional information/features included. These components of DTC Web sites are important to understand and apply to public policy given their potential impact on learning and persuasion.

The majority of the sites (80%) had some kind of human or animal pictured in the site. The most frequent was a woman (81%), followed by a man (70%), child (38%), doctor (29%), animal (14%), "other" (7%), and pharmacist (3%). The "other" category included a baby, nurse, scientist, personified egg, personified stomach, and a personified nose.

To answer this part of RQ1, the characteristics of the DTC Web sites revealed some interesting points. Logically, the people pictured in the Web sites tended to be those who would use the product (women, men, and children). Less common was having doctors and pharmacists pictured, presumably to lend medical credibility to the site. Animals were occasionally pictured in sites for ailments that are often thought to be aggravated by pets (e.g., allergies).

The way in which a message is delivered on the Web is particularly interesting given its capacity for multimedia elements and interactivity. This section will report the message format and degree of multimedia found in the DTC drug Web sites, and the next will discuss interactivity. At the most basic level, 100% of the sites contained at least some copy, and 94% used color to some degree (e.g., font, background, pictures, etc.). The majority of the sites (89%) included at least one picture, which was most commonly color (64%); however, 16% of the sites included both color and black and white pictures, and 9% contained only black and white. Although the "inclusion of graphics" variable had to be dropped because of low reliability, many of the sites included graphics that were as simple as a stylized logo to graphical depictions of complex medical concepts or data, 40% of which contained more than ten different graphics. An impressive 48% of the sample contained at least one animated graphic. In some cases, these were simple animations to catch people's attention and spice up the Web site (e.g., Rhinocort Aqua's butterflies); however, in many instances the animations were being used to demonstrate key medical or treatment information (e.g., Lipitor's "Artery Gallery"). Only 20% of the sites used video to communicate, and 13% used audio. Finally, 14% of the sites had "other" message formats--offering to view the site in Spanish and one that had an option for larger type. This indicates that DTC drug Web sites are using both traditional advertising message characteristics, such as copy and pictures, as well as the multimedia capabilities of the Web to communicate their messages and potentially help consumers learn.

Are the Sites Utilizing the Interactive Capabilities of the Web?

Because multimedia is closely tied to message format, the details about multimedia elements were discussed in the previous sections and found to be rather prevalent. This section focuses on interactivity. To capture the diversity of interactivity, the code sheet offered the coders many options to choose from, including an "other" category. Table 4 summarizes the types and prevalence of interactivity.

As recommended (Coney 2002), links to other sources of information (e.g., nonprofit associations related to condition) were included to a high degree (68%). Other interactive features included a flu report, local pollen count, pollen forecast, ask a physiologist a question, ask a nutritionist a question, interactive nutrition label, calorie burn guide, send in a picture and they will show you how you look with hair, send an e-postcard, tip of the week, food and dosing wheel, customized allergy profile, click and drag a picture as navigation, and neighborhood events.

Both the diversity and degree of interactivity used in these sites was quite high. DTC Web sites are doing a good job of harnessing the power of the Web to educate and form relationships with the consumer by using graphics (e.g., charts, tables, etc.), animation (e.g., how drugs work, how the condition affects the body, etc.), and interactivity (e.g., links, email updates/newsletters, assessment quizzes, etc.).

In the attempt to capture as complete a picture of these Web sites as possible, several other features and pieces of information were recorded that were thought to be key to fully understanding public policy issues of DTC Web sites. Most of the sites (81%) had a statement of the confidentiality or privacy policy, although this information was difficult to find on some sites because it was called "legal information." A copy of advertisements being included in the sites was rare, with only four sites doing this (two television, one radio, and one print). Seven sites showed a cost comparison between the site and other treatment options. Frequently asked questions (FAQs) were found on 66% of the sites, while a site map was only found on 50% of the sites. These are often important features to assist the individual in finding the information he or she needs. Personal testimonials were found fairly frequently on the DTC drug sites (30%). This seems to be an appropriate tool for communication about a prescription drug. Forty-three percent of the sites included information or a form to help the individual ask questions and talk to his or her doctor, which seems low considering this is the only way to obtain a prescription. Finally, it was somewhat surprising to find that contact information was only found on 63% of the Web sites. This is a key piece of information necessary for building a relationship with the consumer. For something as important as a prescription drug and one's health, it is essential to have contact information for questions and concerns.

DISCUSSION

While the results of this study help to illuminate the public policy implications of pharmaceutical companies having DTC drug Web sites, to better understand this type of advertising it is also important to understand how DTC drug Web sites may be the same or different from DTC print ads. This comparison will primarily focus on two studies: Bell, Kravitz, and Wilkes 2000 and Bell, Wilkes, and Kravitz 2000. The present study was not carried out at the same time as these previous studies; therefore, differences in the data may either reflect real differences by medium or could just be the result of differences in coding of the studies (different coders, etc.). These differences in method were limited by using the same coding scheme and definitions, however. Therefore, the differences should truly be due to media differences to a great degree.

The higher level of monetary inducements found on the Web (52%) than in previously researched print DTC ads (17%) (Bell, Kravitz, and Wilkes 2000) indicates that the industry feels that the Web is a better medium to get consumers to take action by requesting rebates and other incentives. A main reason for this may be the ease with which this can be accomplished on the Web (i.e., e-mail or on-line form) as opposed to print (i.e., cutting out and mailing in) or television (i.e., writing down phone number and calling). This is a concern for public policy, however, because of the theorized relationship between inducement and coercion to take a drug without fully weighing the medical options. Sending additional information is another type of inducement. However, the differences between the two media were not as pronounced--47% of the Web sites offered to send additional information, compared to 35% of print ads in the Bell, Kravitz, and Wilkes study (2000).

Overall, there were few differences between the appeals found in previous print studies and those found on the Web in this study, indicating that the sales strategy is simply common to this type of advertising. The differences in advertising appeals used in print and on the Web may indicate what the Web can communicate better or worse than print. A comparison of the Bell, Kravitz, and Wilkes (2000) data to Table 3 reveals that only 7 out of 19 appeals had visually apparent differences, including innovative (41% versus 27%), lifestyle (6% versus 44%), nonmedicated (14% versus 2%), safe (11% versus 24%), convenience (38% versus 52%), quick acting (6% versus 14%), and easy on the system (3% versus 9%). Of these, all were found at a higher rate on the Web than in print, except that innovative and nonmedicated were less common on the Web. The other two print studies affirm the similarities in appeals used in both media. Parker and Delene (1998) found that most appeals were problem/solution, which would include symptom control, effectiveness, convenience, and many others. Similarly, Woloshin et al. (2001) found that improving symptoms and treating the disease were most common. These findings show that DTC drug advertising appeals are very similar on the Web and in print. Counterintuitively, the innovativeness of a drug was used less on the Web than in print. A study by Maddox, Mehta, and Daubek (1997) found that traditional advertising, which included a URL, resulted in the company being perceived as more high-tech. It would seem that a drug being innovative would also be more likely to be communicated in the more high-tech medium of the Web. The higher usage of the lifestyle, safety, convenience, quick acting, and easy-on-the system appeals on the Web indicates more overall focus on medical sorts of appeals on the Web than in print.

A more subjective observation of these sites is that very little of the content was aimed at advertising appeals; rather, the majority seemed to be educational in nature. This is supported by the measurement of medical information. The results of this study showed that DTC drug sites included a great deal more medical and drug information than did the previous print study (breadth and depth). A comparison of Bell, Wilkes, and Kravitz's (2000) findings (presented in the the Literature Review section) and Table 2 reveals every category of medical information had a lower percentage represented in the previous study's sample of print ads than was found in the current study's DTC Web sites, in many cases more than twice as high (condition name, 95% versus 97%; precursors, 27% versus 74%; prevalence, 12% versus 66%; misconceptions, 9% versus 21%; mechanism of action, 36% versus 83%; supportive behaviors, 24% versus 69%; alternative treatments, 29% versus 63%; time to onset of action, 20% versus 52%; treatment duration, 11% versus 50%; and success rate, 9% versus 23%). The only piece of information that was not very different was the condition name. Granted, the amount of information that the Web can convey is exponentially higher than a print advertisement. However, this may indicate an inherent strength of the Web to convey more detailed information than print communication can. Consumers have more time to read and digest the information, the medium can better organize the quantity of information required for pharmaceutical drugs, and the Web may even be able to better teach consumers the technical aspects of drugs and the conditions they treat (as discussed in future research). Therefore, the Web is better able to meet the FDA requirement of "adequate provision" by "communicating all information relevant to the product's indication (including limitations to use) in consumer-friendly language" (FDA 1999). This research supports the important role that DTC Web sites play in the public policy of drug campaigns.

The format of Web DTC sites also makes conveying this large amount of technical information better suited to the Web as opposed to print or television. The Web is able to uniquely combine much of what print (e.g., text and graphics) and television (e.g., audio and video) has to offer while adding the additional advantage of interactivity, which is not possible to the same degree in print or on television. This study found many different types of interactive features included in the Web site that, based on previous empirical research (Haubl and Trifts 2000; Li and Bukovac 1999; Macias 2003), may increase consumers' comprehension as well as their involvement.

CONCLUSIONS

This research has added a key component to the existing research dealing with public policy and DTC advertising. Compared with previous print studies, the results of this research indicate that the format and content of DTC drug Web sites may be better able to meet FDA guidelines. Specifically, the sites include more medical and drug information (including risks and limitations) as opposed to more selling messages, and they have interactive and multimedia capabilities that may lead to improved learning. To summarize, this study shows strong support that DTC Web sites could be an important source of information for consumers with medical concerns related to DTC drugs. While print and television can do an excellent job of making consumers aware of a drug, the Web may be better suited for educating the consumer because of its demonstrated ability to include more medical information. However, the increased use of inducements should be a concern for policymakers. The major limitation of this study is that it can only describe what DTC drug Web sites are communicating and not what effect they may have on consumers. Therefore, many of the suggestions for future research deal with extensions into this area because it is the next important step.

As mentioned earlier, this study added to a previous typology (Bell, Wilkes, and Kravitz 2000) and presented a complete list of medical information content used in advertising. This is an important issue, as medical information is taking a more prominent position in consumers' minds with the explosion of DTC advertising, the increasing cost of health care to consumers, and the aging baby boomer generation. Future research could apply this more complete typology to print, television, and direct mail.

The DTC Web sites are clearly focusing on consumers' needs for the easy relief that the advertisements for the drugs promoted. However, future research should investigate what effect the informational content of the DTC drug Web sites has on consumers' attitudes. Are these sites encouraging positive attitudes and the potential for repeat visits because they are rich in information? The findings show that, overall, appeals are very similar on the Web and in print. The important differences emerged in the areas of message format and medical information.

The DTC Web sites are characterized by a relatively high degree of multimedia elements and interactivity. Previous studies of a variety of Web sites (e.g., television stations, commercial products and services, etc.) have found that they do not fully utilize the interactive, multimedia, and personalization aspects of the Web (Chan-Olmsted and Suk Park 2000; Frazer and McMillan 1999). Part of the reason for this may be the type of industry and product--this is a caregiving industry that strives to help people, and prescription drugs are most often very high-involvement products for which consumers need detailed information. Another reason for this is to try to establish a competitive advantage by forming relationships with brand-loyal customers. This is essential because DTC brands are limited in their ability to use comparative advertising techniques to persuade customers that their brand is better than those of their competitors. It is interesting to note, however, that some brands may focus on slight product differentiations (e.g, indoor and outdoor allergy relief) while not making direct comparisons to the competitor. It is essential for public policy that future research provide better understanding about the degree to which consumers are even visiting DTC drug sites. What are the effects of multimedia elements and interactivity in DTC drug Web sites? Do they lead to increased learning and, if so, when? Or do they simply serve as a marketing tool?

Future research also needs to focus on understanding whether consumers are getting the information they need about prescription drugs in a manner they can fully understand. The Web seems to be a particularly good medium for this type of comprehension, as indicated by the higher level of information, particularly medical information, presented in DTC Web sites as compared with print ads. The consumer can take as much time as he or she needs to fully understand the detailed and often complicated information. Some of the interactive features, such as Lipitor's animated "Artery Gallery," are being used to demonstrate medical or treatment information in a way that is more comprehendible for the layperson. However, the issue of whether this is a more educationally effective way to communicate information found in the brief summary should be a focus of future consumer research.

Finally, this study has implications for integrated marketing communications. Regarding the ease of finding a DTC site's URL, it seems that traditional DTC ads can, and often do, play a key role. As was mentioned earlier, although several of the sites' URLs are www.brandname.com, it is often difficult to find these sites because the name is hard to spell or otherwise not commonly known (e.g., Xeloda or Xenical). However, even among those not in the market for the drug, few are unfamiliar with the names of those who spend the most on advertising--Claritin, Allegra, Viagra, and Lipitor. Zyrtec seems to understand the importance of brand-name recognition, and even makes efforts to educate consumers about how to pronounce and spell the brand name. Many of its advertisements included the following instructions: "Learning how to say ZYRTEC (zur'-tek) is the hard part. Getting allergy symptom relief can be easy." An integrated marketing approach also has public policy implications, because it may increase the number of consumers who get the information they need to make an educated decision. Finally, as was indicated by the earlier listing of DTC expenditures, the pharmaceutical and advertising industries may not agree that DTC Web sites are "advertising." It would be interesting to conduct in-depth interviews with marketers and advertisers of DTC drugs to further explore this question and better understand their communication purpose.

TABLE I
Medical Conditions and Web Sites

Medical condition        Frequency   Brands

Psychiatric/neuro-          14       Aricept, Buspar, Celexa,
  logical disorders                    Effexor XR, Imitrex, Paxil,
                                       Prozac, Risperdal, Serzone,
                                       Topamax, Wellbutrin, Zoloft,
                                       Zomig, Zyprexa
OB/GYN conditions           10       Alesse, Depro-Provera,
                                       Diflucan, Ortho-Prefest,
                                       Ortho-Tricyclen, Paragard,
                                       Premarin, Preven, Terazol,
                                       Vagifem
Other                       10       Ambien, Hydrocodone, Meridia,
                                       Oxandrin, Propecia, Synthroid,
                                       Transdermal, Xalatan, Xenical,
                                       Amaryl (condition not listed)
Musculoskeletal              8       Celebrex, Daypro, Evista,
  ailments                             Fosamax, Miacalcin, Tasmar,
                                       Ultram, Vioxx
Allergies                    7       Allegra, Claritin, Flonase,
                                       Nasonex, Rhinocort, Rhinocort
                                       Aqua, Zyrtec
Cardiovascular disease       7       Altace, Coumadin, Lipitor,
                                       Plavix, Pravachol, Procrit,
                                       Zocor
HIV/AIDS                     7       Combivir, Crixivan, Fortovase,
                                       Marinol, Viracept, Viramune,
                                       Zerit
Respiratory conditions       6       Accolate, Azmacort, Combivent,
                                       Levaquin, Pulmicort Respules,
                                       Pulmicort Turbohaler
Dermatological               5       Differin, Lamisil, Metrogel,
  conditions                           Soriatane, Sporanox
Infectious/non-HIV           5       Denavir, Famvir, Lariam,
  diseases                             Tamiflu, Valtrex
Urological conditions        3       DetrolLA, Proscar, Viagra
Cancer                       2       Nolvadex, Xeloda
Diabetes                     2       Glucophage, Regranex
Gastrointestinal             2       Nexium, Prevacid
  conditions
Tobacco addiction            2       Nicotrol, Zyban

Note: n = 90.

TABLE 2
Medical Information in Direct-to-Consumer (DTC) Web Sites

                                DTC drug Web sites

Medical condition information
  Condition name                       97%
  Symptoms                         Dropped (low
                                     reliability)
  Precursors                           74%
  Prevalence                           66%
  Misconceptions                       21%

Treatment information
  Side effects                         94%
  Prescribing information              93%
  Mechanism of action                  83%
  Contraindications discussed          80%
    in Web site content
  Supportive behaviors                 69%
  Alternative treatments               63%
  Time to onset of action              52%
  Treatment duration                   50%
  Success rate                         23%

Note: Percentages do not add up to 100 because each
site could potentially have numerous pieces of
information.

TABLE 3
Advertising Appeals and Selling Points in Direct-to-Consumer
(DTC) Web Sites

                                    DTC drug Web sites

Effectiveness
  Effective                                63%
  Symptom control                      Dropped (low
                                         reliability)
  Innovative                               27%
  Prevention                               22%
  Reduced mortality                        10%
  Cure                                      6%
  Powerful                                  6%
  Dependable                                3%

Ease of use
  Convenience                              52%
  Quick acting                             14%
  Easy on system                            9%
  Economical                                6%

Social-psychological enhancements
  Lifestyle                                44%
  Psychological                            16%
  Social                                    6%

Safety
  Safe                                     24%
  Nonaddictive                              8%
  Natural                                   4%
  Nonmedicated                              2%

Note: Percentages do not add up to 100 because each site
could potentially use numerous appeals.

TABLE 4
Prevalence of Interactive Features

Interactive feature                                Percent

Clickable links within site or to other sites        92%
Link to manufacturer's homepage                  Dropped (low
                                                   reliability)
Links to nonprofit related to condition              68%
Register                                             62%
Search feature within site                           38%
Questionnaire/survey                                 29%
E-mail friend                                        27%
Submit a question                                    21%
Refill reminder                                      18%
Assessment quiz/survey/checklist                     14%
Feedback                                             11%
Diary (symptom, eating, etc.)                        10%
Search engine to locate medical personnel            10%
Get/exchange recipes                                  5%
BMI calculator                                        4%
Glossary                                              4%
"Perks program" (frequent buyer)                      3%
Games                                                 3%
Bookmark                                              2%
Get site updates e-mailed                             2%
Personalize/customize site                            1%
Chat room                                             0%

Note: n = 90.

REFERENCES

Alperstein, Neil, and Mark Peyrot (1993), "Consumer Awareness of Prescription Drug Advertising," Journal of Advertising Research (July/August), 50-56.

American Marketing Association (AMA) (1997), Policy Compendium, 227-228.

Bell, Robert A., Richard L. Kravitz, and Michael S. Wilkes (2000), "Direct-to-Consumer Prescription Drug Advertising, 1989-1998: A Content Analysis of Conditions, Targets, Inducements, and Appeals," Journal of Family Practice, 49 (4), 329-335.

--, Michael S. Wilkes, and Richard L. Kravitz (2000), "The Educational Value of Consumer-Targeted Prescription Drug Print Advertising," Journal of Family Practice, 49 (12), 1092-1098.

Burton, Philip Ward (1990), Advertising Copywriting, 6th ed., Lincolnwood, IL: NTC Business Books.

Calfee, John E. (2002), "Public Policy Issues in Direct-to-Consumer Advertising of Prescription Drugs," Journal of Public Policy and Marketing, 21 (2), 174-193.

Chan-Olmsted, Sylvia, and Jung Suk Park (2000), "From On-Air to Online World: Examining the Content and Structures of Broadcast TV Stations' Web Sites," Journalism and Mass Communication Quarterly, 77 (2), 321-339.

Cohen, Eric P. (1998), "Sounding Board: Direct-to-Consumer Public Advertisement of Prescription Drugs," New England Journal of Medicine, 373-375.

Coney, Sandra (2002), "Direct-to-Consumer Advertising of Prescription Pharmaceuticals: A Consumer Perspective from New Zealand," Journal of Public Policy and Marketing, 21 (2), 213-223.

Coulter, Angela, Vikki Entwhistle, and David Gilbert (1998), Informing Patients: An Assessment of the Quality of Patient Information Materials, London: King's Fund.

Coyle, James R., and Esther Thorson (2001), "The Effects of Progressive Levels of Interactivity and Vividness in Web Marketing Sites," Journal of Advertising, 30 (3), 65-77.

Federal Trade Commission (FTC) (1996), "Before the Department of Health and Human Services Food and Drug Administration in the Matter of Direct-to-Consumer Promotion" (January 11, Public Hearing Docket no. 95N-0227), available at www.ftc.gov (accessed June 3, 2003).

Fetto, John (2002), "Drugged Out," American Demographics, 24 (5), 11.

Food and Drug Administration (FDA), Center for Drug Evaluation and Research (CDER) (1997), "Draft Guidance for Industry: Consumer-Directed Broadcast Advertisements," Federal Register, 62, no. 155 (August 12, Docket no. 97D-0320), 43171-43173.

--(1999), "Guidance for Industry: Consumer-Directed Broadcast Advertisements," Federal Register, 64, no. 152 (August 9, Docket no. 97D-0302), 43197-43198.

Frazer, Charles, and Sally J. McMillan (1999), "Sophistication on the World Wide Web: Evaluating Structure, Function, and Commercial Goals of Web Sites," in Advertising and the World Wide Web, David W. Schumann and Esther Thorson, eds., Mahwah, NJ: Lawrence Erlbaum, 119-134.

Friedler, Edward (1997), "The Evolving Doctor-Patient to Provider-Consumer Relationship," Journal of Family Practice, 45 (December), 485.

Green, K. (1998), "Marketing Health-Care Products on the Internet: A Proposal for Updated Federal Regulations," American Journal of Law and Medicine, 24 (2/3), 365-386.

Haubl, Gerald, and Valerie Trifts (2000), "Consumer Decision Making in Online Shopping Environments: The Effects of Interactive Decision Aids," Marketing Science, 19 (1), 4-21.

The Henry J. Kaiser Family Foundation (2001), "Understanding the Effects of Direct-to-Consumer Prescription Drug Advertising" (November).

Holsti, Ole R. (1969), Content Analysis for the Social Sciences and Humanities, Reading, MA: Addison-Wesley.

Humphreys, Andrew (2000), "Out of Site: The Top 10 HealthCare Websites Are Fast, Informative, Artistic, Innovative, Creative, and Easy to Use," MedAdNews, 19 (12), 5, 10-12.

Kelley, Gordon (2002), "Brand Advertising on a Roll," Pharmaceutical Executive, 22 (6), 118.

Krippendorff, Klaus (1980), Content Analysis: An Introduction to Its Methodology, Beverly Hills, CA: Sage.

Law, Jacky (1998), "Assessing the Impact of Direct-to-Consumer Advertising," Scrip Magazine (November), 21-22.

Li, Hairong, and Janice L. Bukovac (1999), "Cognitive Impact of Banner Ad Characteristics: An Experimental Study," Journalism and Mass Communication Quarterly, 76 (2), 341-353.

McCarthy, Robert (1998), "Managed Care Matters--Major Managed Pharmacy Issues: Cost, Cost ... Copays?" Drug Benefit Trends, 10 (9), 22, available at www.medscape.com (accessed June 3, 2003).

Macias, Wendy (2003), "A Beginning Look at the Effects of Interactivity, Product Involvement and Web Experience on Comprehension: Brand Web Sites as Interactive Advertising," Journal of Current Issues and Research in Advertising, 25 (2), 31-44.

McMillan, Sally J. (2000), "The Microscope and the Moving Target: The Challenge of Applying Content Analysis to the World Wide Web," Journalism and Mass Communication Quarterly, 77 (1), 80-98.

--, and Jang-Sun Hwang (2002), "Measures of Perceived Interactivity: An Exploration of the Role of Direction of Communication, User Control, and Time in Shaping Perceptions of Interactivity," Journal of Advertising, 31 (3), 41-54.

Maddox, Lynda M. (1999), "The Use of Pharmaceutical Web Sites for Prescription Drug Information and Product Requests," Journal of Product and Brand Management, 8 (6), 488-501.

--, Darshan Mehta, and Hugh G. Daubek (1997), "The Role and Effect of Web Addresses in Advertising," Journal of Advertising Research (March/April), 47-59.

Matthews, Merrill (2001), "Advertising Drugs is Good for Patients: Lowers Costs to Consumers," Consumers' Research (August), 15-20.

Nader, Ralph (1999), "Is DTC in the Best Interest of Public Health?" in IBC's 1999 Inaugural DTC Marketing Congress, Sothborough, MA: IBC.

Neff, Jack (2002), "Drugmakers Up Use of Celebs in Ads, Risking an Overdose or Critics' Ire," Advertising Age (May 24, Special Report: DTC Marketing), S-4.

Novak, Thomas P., Donna L. Hoffman, and Yiu-Fai Yung (2000), "Measuring the Customer Experience in Online Environments: A Structural Modeling Approach," Marketing Science, 19 (1), 22-42.

Parker, Betty J., and Linda M. Delene (1998), "Direct-to-Consumer Prescription Drug Advertising: Content Analysis and Public Policy Implications," Journal of Pharmaceutical Marketing and Management, 12 (4), 27-42.

--, and --(1999), "The Marketing of Direct-to-Consumer Prescription Drugs: An Examination of Advertising Content," Journal of Promotion Management, 5 (1), 37-55.

People's Medical Society (1996), "Drug Ads to Consumers on the Rise," People's Medical Society Newsletter, 15 (April), 1.

Peterson, Carolyn (1998), "I Read It over Breakfast: Print Proves Its Appeal with Some All-Ink DTC Campaigns," Pharmaceutical Executive: DTC Times (September 6-10).

Prescription Access Litigation (2002), "Claritin" (September 15), available at www.prescriptionacesslitigation.org/clartin.htm (accessed June 3, 2003).

Realto, Vern (1998a), "Prescription Meds Meet the Web," Pharmaceutical Executive, 18 (11), DTC6-DTC 16.

--(1998b), "1998 Top 10: Best of the Best," Pharmaceutical Executive, 18 (12), S6-S11.

Resnik, Alan, and Bruce L. Stern (1977), "An Analysis of Information Content in Television Advertising," Journal of Marketing, 41, 50-53.

Riffe, Daniel, Stephen Lacy, and Frederick Fico (1998), Analyzing Media Messages: Using Quantitative Content Analysis in Research, Mahwah, NJ: Lawrence Erlbaum.

Roth, Martin S. (1996), "Patterns in Direct-to-Consumer Prescription Drug Print Advertising and Their Public Policy and Marketing Implications," Journal of Public Policy and Marketing, 15 (Spring), 63.

Russell, J. Thomas, and W. Ron Lane (2002), Kleppner's Advertising Procedure, 15th ed., Upper Saddle River, NJ: Prentice Hall.

Scussa, Frank (2001), "Getting Noticed: The Future of Consumer Promotion Is Being Challenged by Government Agencies and the Public," MedAdNews, 20 (6), 1, 40-45.

Twibell, Dan, and Mark Anzalone (2001), "DTC on the Web--Fad or Fundamental?" DTC Perspectives (May/June), 28-29.

Von Knoop, Carina, Deborah Lovich, Martin B. Silverstein, and Michael Tutty (The Boston Consulting Group) (2003), "Vital Signs: E-Health in the United States" (January 3), available at www.bcg.com/publications/files/Vital_ Signs_Rpt_Jan03.pdf (accessed June 3, 2003).

Weare, Christopher, and Wan-Ying Lin (2000), "Content Analysis of the World Wide Web: Opportunities and Challenges," Social Science Computer Review, 18 (3), 272-292.

Wilkes, Michael S., Robert A. Bell, and Richard L. Kravitz (2000), "Direct-to-Consumer Prescription Drug Advertising: Trends, Impact, and Implications," Health Affair, 19 (2), 110-128.

Wogalter, Michael S., Tonya L. Smith-Jackson, Brian J. Mills, and Corrina S. Paine (2002), "The Effects of Print Format in Direct-to-Consumer Prescription Drug Advertisements on Risk Knowledge and Preference," Drug Information Journal, 36, 693-705.

Woloshin, Steven, Lisa M. Schwartz, Jennifer Tremmel, and H. Gilbert Welch (2001), "Direct-to-Consumer Advertisements for Prescription Drugs: What Are Americans Being Sold?" The Lancet, 358 (October), 1141-1146.

Wendy Macias (Ph.D., University of Texas at Austin) is an assistant professor in the Department of Advertising and Public Relations, Grady College of Journalism and Mass Communication, University of Georgia.

Liza Stavchansky Lewis (Ph.D., University of Texas at Austin) is a consultant at Spectra Marketing, Kenilworth, NJ.