Then and Now: Examining How Consumer Communication and Attitudes of Direct-to-Consumer Pharmaceutical Advertising Have Changed in the Last Decade

Abstract

This study examines changes over a 10-year period in consumer reports of communication with health care providers about direct-to-consumer advertised (DTCA) medications. Two rounds of survey data were collected in 2003 and 2012 using repeated cross-sectional procedures to examine consumer willingness to discuss DTCA medications, content and tone of those conversations, and attitudes about the advertisements. In total, 472 surveys were analyzed. Generally, we found physician–patient conversations, attitudes, and behaviors regarding DTCA have changed. Consumers in 2012 reported talking significantly less about the names of the advertised drug, comparing the advertised drug with their current medication, and sharing general information than consumers in 2003. Attitudes toward the advertisements were significantly more negative in 2012 compared to 2003. Of those who specifically asked for a prescription, the proportion of patients who received the prescription was significantly lower in 2012, despite research suggesting increased rates of prescriptions. These results are interpreted in light of previous research about the lack of research examining the actual communication between physicians and patients on this topic. Limitations of the study are provided along with directions for future research about DTCA and physician–patient communication.

Notes

1. ¹Students were instructed to notify participants that course credit was not based on their completing the survey, in order to reduce the likelihood of participants completing the study under duress of the student’s grade and to lessen any chance of students fabricating results. Students were also given the opportunity to receive points by completing another assignment in the case where surveys were not returned. Related, in the 2012 sample, a small number of upper division and graduate students at one of the three institutions completed surveys. Due to the anonymity of respondents we do not know exactly how many, but we believe it to be approximately 8% of the total 2012 participants.

2. ²Length of relationship with physician has been identified as significant in understanding variations in provider–patient interaction. Due to a slight difference in instructions on the 2003 survey, only those who reported having a DTCA conversation with their provider answered questions about length of relationship (68 of the 246). A change was made in the 2012 survey so all participants answered the question. Thus, length of relationship with physician was not a variable on which we could compare the two groups. However, when examining only those who discussed a prescription medication, we found no significant relationship between the two groups on the variables of physician–patient relationship or reason for the visit.

3. ³Frequencies of those who answered yes on “did you discuss an advertised medication” do not match those on “did you ask” or “did the doctor agree to prescribe.” There is a difference of four people in 2003, and six people in 2012 who reported “I did not ask” but checked yes, “the doctor agreed to prescribe.” While these numbers are not large, they deserve explanation. In both samples, more people answered the latter two questions than the former. We believe participants may not have discussed an ad in the last 6 months, but did recall a conversation with their doctor and responded to the additional questions with that conversation with mind. Also, the reasons that numbers do no match between “did you ask” and “did the doctor prescribe” is that even if someone said they did not ask, there were cases where the participant reported the doctor did prescribe. We suspect that even though they did not directly ask, the fact that the patient brought up an advertised medication led to the doctor being aware of the patient’s interest in it and the doctor offered to prescribe. This is confirmed by comments written by participants who said they did not ask, but the doctor did prescribe. Feng et al. (2011) also reported similar findings that doctors assumed that if patients brought up a medication, they had researched it and thus wanted to try it. Our tests reflect frequencies based on those who report physician-prescribing behavior.

4. ⁴The n values are too small to run a meaningful Z score on the difference in refusals to prescribe.