Lack of Impact of Direct-to-Consumer Advertising on the Physician-Patient Encounter in Primary Care: A SNOCAP Report

Dr Parnes and colleagues’ study makes an important contribution to the available research evidence on the effects of direct-to-consumer advertising (DTCA) on prescribing. [1] However, to call the results a ‘lack of impact’, as stated in the title, is highly misleading.

In Parnes’ study, if a patient requested a medicine, the physician prescribed it 54% of the time (n=31), but physicians only considered 12/31 (39%) of these medicines to be first choice treatments for the patient. Similarly, in a study of DTCA in Sacramento and Vancouver, we found that physicians only judged half of the requested advertised drugs they prescribed to be very likely treatment choices for other similar patients, as compared with 88% of non-requested drugs. [2] Both results suggest a degree of physician ambivalence about treatment choice.

Secondly, Parnes et al. found that physicians felt ‘somewhat’ or ‘significantly’ pressured in a third of consultations in which a patient had requested a medicine. In the absence of comparative information, the authors consider this reassuring. In our study, physicians reported some degree of pressure in 15 (20%) of 74 consultations involving DTCA drug requests. [2] However, in total they only reported pressure in 48 (3%) of 1431 consultations, including these 15. The direction of effect – towards more reported pressure – is unmistakable.

In the Sacramento arm of our study, 7.2% of patients requested advertised medicines, a subset of the 15.8% requesting one or more medicine. [2] Parnes reports fewer patient requests: 3.5% overall and 7.2% in private offices, around two-thirds for advertised medicines. Our population was more similar to the private than clinic patients. We included only English-speaking adults, since patients filled in a questionnaire. Parnes’ patients were 44% Hispanic, many non English- speaking, and 24% under 18. These groups likely have lower DTCA exposure. In our study, patients with less self-reported exposure were less likely to request an advertised medicine.

Patients in Parnes’ study who could not afford advertised drugs may also have requested fewer. Hansen et al. found that insurance coverage and co-payments affected the likelihood that a patient switched to an advertised proton pump inhibitor. [3]

Our results and Parnes’ are in many ways qualitatively similar, yet the interpretation differs dramatically. As Peter Cronholm and colleagues point out, the magnitude of effect on prescribing in Parnes’ study is consistent with billions of dollars in extra sales each year. [4] Is this truly a lack of impact? And if DTCA leads to more prescriptions despite physician ambivalence, are patients protected from another rofecoxib (Vioxx)-like disaster?

References
1. Parnes B, Smith PC, Gilroy C et al. Lack of effect of direct-to- consumer advertising on the physician-patient encounter in primary care: a SNOCAP report. Ann Fam Med 2009; 7(1): 41-46.
2. Mintzes B, Barer ML, Kravitz RL, Bassett K, Lexchin J, Kazanjian A, Evans RG, Pan R, Marion SA. How does direct-to-consumer advertising (DTCA) affect prescribing? A survey in primary care environments with and without legal DTCA. CMAJ 2003 169: 405-412.
3. Hansen RA, Shaheen NJ, Schommer. JC. Factors influencing the shift of patients from one proton pump inhibitor to another: the effect of direct-to-consumer advertising. Clin Therapeut 2005; 27: 1478-1487
4. Cronholm P, The health service burden of “low impact” direct to consumer advertising. [e-letter] Ann Fam Med, 22 Jan 2009. Available at: http://www.annfammed.org /cgi/eletters/7/1/41#10517

Competing interests:   None declared

While reassuring that the impact of Direct to Consumer Advertizing (DTCA) described in Dr. Parnes’ article was lower than in previous estimates, even the conservative numbers presented have a significant impact on society when framed using a population perspective. With approximately 250 billion dollars spent on pharmaceutically-related healthcare expenditures estimated for 2008, the most conservative estimate one could calculate based on half of 3.5% of visits resulting in the use of a DTCA medication would be close to 4.5 billion dollars [1]. The rate doubles for patients seen in private offices where 82% of outpatient visits take place [2]. As DTCA intentionally focuses on the most expensive drugs available, the true economic impact would likely represent an order of magnitude or more within the current US healthcare system and its regretfully low return on investment. We remain skeptical that DTCA has a “low impact” on healthcare and concerned that almost two-thirds of providers in the study reporting regular contact with the pharmaceutical sales industry.

We also express some concern about the increased time burden on healthcare providers that may result from DTCA. While this report suggests that patient request may have the “benefit” of increased patient education and discussion, the specific characteristics of such discussions have not been described. For example, DTCA-prompted medication requests might require providers to educate patients about why the requested medications are not necessarily the best choice; consequently, DTCA may be generating a need for providers to correct misconceptions and supplement incomplete marketing messages. If 37.9% of visits are lengthened as a result of DTCA, a provider in a solo private practice would potentially need to do this type of counseling about 200 times per year (37.9% of visits x 7.2% of patients x 30 patients/day x 5 days/week x 48 weeks). In an era when primary care physicians do not even have adequate time in a typical workday to provide age-appropriate preventive services, the need to spend added time for such conversations may lead to further displacement of medically important services [3].

[1] Centers for Medicare and Medicaid Services. <http://www.cms.hhs.gov/NationalHealthExpendData/25_NHE_Fact_Sheet.asp#TopOfPage >. Accessed 2009 January 16.

[2] Fairbrother G, Gusmano MK, Park HL, Scheinmann R. Care For The Uninsured In General Internists' Private Offices. Health Aff. 2003 November 1, 2003;22(6):217-24.

[3] Yarnall, K.S., et al., Primary care: is there enough time for prevention? Am J Public Health, 2003. 93(4): p. 635-41.

Competing interests:   None declared

Thanks much to both of you for your thoughtful comments. This is indeed a topic that generates strong opinions. As an aside, I would like to note that the idea for this project came from a practicing clinician in our PBRN--this is clearly not just a topic for academics.

Dr Allen, I agree with your thoughts about the trend toward advertising diseases and symptoms. Suppose a company has a large market share for a certain condition, such as BPH, then advertises the drug in order to gain market share from a competitor. This will probably not lead to as many new prescriptions compared to getting new patients with the condition in for treatment. These new patients are likely to be prescribed the company's product because of its large market share. Another advantage of this approach is that when a condition is advertised, there is no requirement to state all the unpleasant side effects that may occur.

I am pleased to hear that you, like many of the clinicians from our survey, have found a way to make patient medication inquiries into a positive experience for the patient's health. Related to this, given the vast amount of DTCA, clinicians and patients are accustomed to discussing it in clinical encounters, compared to several years ago when patient inquiries may have caused more clinician discomfort.

Dr Brody, perhaps the only outlet for clinicians to express their discontent is in seminars like the ones you present. Your suggestion for a focus group follow up project is a good one. Point of care surveys by necessity need to be very brief, and this is both a strength in obtaining a large sample size and limiting recall bias, and a limitation in the lack of detail that can be obtained.

Competing interests:   None declared

Compare these two phenomena:

On the one hand is the experience reported in this paper. It is very rare that physicians encounter a patient seeking a drug based on DTCA. When it happens, the physicians seem to regard the event as neutral or positive within their relationship with the patient.

On the other hand is the anecdotal experience I can report as a speaker who commonly talks about the ethics of the relationship between us and the drug industry. I generally do not mention DTCA in the main talk as it is peripheral to the question of OUR relationship with Pharma, and whether we act in trustworthy ways in that relationship. It seems that the first, second or third question from the audience will be about DTCA and will suggest a criticism of the practice, with the suggestion that I did not tell the whole story of the drug industry's misdeeds because I did not condemn DTCA.

If DTCA is no big deal in this practice-based study where encounters were tracked in real time, why do so many of us think it is such a big deal when the topic is raised in the abstract? What is it about DTCA that grinds on us as much as it does?

One area that this study just hints at is the interesting fact that we say that the impact is either neutral or positive; that we say we feel little pressure to prescribe the drug--but we then proceed to prescribe the drug the majority of the time. So something here does not match up, and maybe the feelings that stirs up are more intense than the surface impression. Anyone for a follow-up round of focus group research?

Competing interests:   None declared