Facebook and Pharmaceutical Companies: An Industry in Need of Guidance

Companies across product categories are working to learn the nuances of social media as it continues to grow in popularity and wide spread adoption. Pharmaceutical companies are no different, except that they must also adapt government regulation and industry standards to the new medium, making the topic important for pharmaceutical companies, consumers, and policy makers. The paper outlines the use of one specific social media platform, Facebook, by three of the world's largest pharmaceutical companies. In addition the paper includes an exploratory crossover study that examines consumer perceptions of the content on the different sites with regard to informative value, entertainment value, and learning. Results indicate that consumers perceive differences on all three variables. Open-ended responses are used to discuss some potential explanations for these differences.

0.94 for entertainment. Learning was measured with a single item. Scale items and reliabilities are provided in table 1.

Results
Repeated measures ANOVA was used to look for mean differences across the Facebook sites on the variables of interest (informative value, entertainment value and learning. The results indicate perceptual differences do exist. Table 2 shows the means for the conditions of the design. With regards to RQ1 and informative value, the Pfizer site ranked the highest on this variable ranking significantly better than both Novartis (α < .01) and AstraZeneca (α < .01). AstraZeneca ranked second on this variable, scoring significantly higher than Novartis (α < .01). RQ2 focused on the entertainment value of the Facebook site. Again Pfizer had the highest mean on entertainment value and was perceived as more entertaining than the Novartis page (α < .01).
AstraZeneca was also perceived as more entertaining than Novartis (α < .01). There were not significant differences between Pfizer and AstraZeneca (α = .203).
RQ3 concentrated on how much consumers believe that they are learning from a Facebook page.
Consumers associated the greatest learning with the Pfizer Facebook page. Respondents perceived significantly more learning from Pfizer than Novartis (α < .01) or AstraZeneca (α < .01), but no significant difference between Pfizer and AstraZeneca (α = .065) or AstraZeneca and Novartis (α = .249). Pairwise comparisons across the groups are shown in table 3. To test RQ4, a paired-samples t-test was performed to look for differences in the mean ranking score for each of the Facebook sites. The means for ranking are listed in table 4 (lower numbers indicate a higher ranking). Pfizer had the best average ranking of the 3 Facebook sites followed by Novartis, and AstraZeneca ranked worst. For policy makers, these exploratory results reinforce that consumers are likely to perceive differences in informativeness and entertainment. Both are variables of concern from a regulatory perspective. As with other types of media, guidelines need to consider balance. Once pharmaceutical companies take the plunge into offering drug information via Facebook, regulations will have to go beyond the balance of positive and negative information. They will need to consider the balance of information versus entertainment content. It is no wonder that the FDA has refused to even put a time frame on when it will offer guidance after the most recent delay (Schwartz 2011). It is a complicated media and a comprehensive process.
As with any research, the study in this paper has some limitations. First of all, the study is exploratory. It is only an initial investigation of a wide range of potential variables and situations that can and should be studied. As such, it makes no attempt at proposing or testing causation nor does it consider every variable that may influence persuasion, but it does offer some interesting initial insights on a topic of both theoretical and practical importance. Second, the use of a student sample may raise concerns about the generalizability of the results. Despite conflicting opinions on the use of student subjects, they are generally acceptable for theory testing and more exploratory research (Bergmann and Grahn 1997).