RESEARCH LETTER: Do Physicians Discuss Political Issues with Their Patients?

To the Editor: Politicians have an important influence on the structure and function of our health care system. As a result, it seems likely that physicians and their patients might discuss political issues of consequence to health care. Yet, little is known about physicians' attitudes and activities in this regard or about their level of personal political participation.^1,2,3,4 We conducted a survey to examine the political attitudes and activities of physicians and to identify characteristics associated with physician willingness to address political issues with clinic patients. We developed a 23-item, 1-page questionnaire and pilot tested it with 8 physicians. The instrument measured personal political activities, political activities in the clinic, and attitudes about addressing political issues with clinic patients. Excluding demographics, response options were dichotomous (yes/no) or 5-point Likert scales. The University of Pennsylvania Institutional Review Board approved the study, which was conducted after Pennsylvania's voter registration deadline but before Election Day (October 5 to November 1, 2004). All primary care physicians (PCPs) in the Division of General Internal Medicine practicing in the University of Pennsylvania Health System were asked to complete the questionnaire. Six practices in urban and suburban Philadelphia were surveyed. Hospitalists and clinician-researchers were excluded. Participation was optional and consent implied. Questionnaires were self-administered and anonymous. For analyses, Likert responses were dichotomized (agreement/disagreement) and “unsure” responses were excluded. Bivariable associations were examined with χ² or Fisher's exact tests as appropriate using STATA 8.0 (Stata Corp., College Station, Tex). Of the 37 eligible PCPs, 36 (97%) participated. Study PCPs practiced clinical medicine for medians of 9 years (25% to 75% interquartile range (IQR) 7 to 15) and 32 hours per week (IQR 20 to 40), and 19 (53%) were male. All PCPs had indigent patients but 6 (17%) primarily served patients with annual household incomes less than $25,000. Table 1 summarizes our survey findings. Overall, 34 PCPs (94%) discussed voting or a politically oriented health care issue with a patient and 20 (56%) initiated such discussions. Twenty-two (63%) were willing to offer voter registration in their clinics, while 30 (83%) were willing to offer information on voter registration. Thirty-one (86%) agreed that political communication with patients, either through brochures or direct discussion, was appropriate. Male PCPs were more likely to contribute money to political campaigns (15 of 19 vs 8 of 17, P =.05) and to initiate discussions with patients about politically oriented health care issues (12 of 19 vs 5 of 17, P =.04). Physicians who contributed money or volunteered time to political campaigns were more likely to support voter registration in the clinic (18 of 23 vs 4 of 12, P =.02). PCPs who primarily served low-income patients were less likely to regard discussions about politically oriented health care issues as appropriate (0 of 5 vs 13 of 22, P =.04). In conclusion, the majority of our study PCPs discussed voting or health care politics with their patients. To our knowledge, this is the first study to demonstrate that the personal political activities of physicians are associated with their willingness to address political issues with clinic patients. Moreover, we found that gender was associated with political activity inside and outside the clinic, a finding consistent with a study of English PCPs.⁵ Our study was limited by the small sample size, the single health care system source of our cohort, unadjusted analyses, and self-report. Future studies should continue to explore the political attitudes and activities of physicians in different populations and settings, in addition to addressing the influence of such activities on the patients they serve. Moreover, we believe that a debate on the ethical issues surrounding political discourse in the practice setting is warranted.—

Table 1.

Political Activities and Attitudes of UPHS Clinician Cohort (N=36)

Variables	N (%)
Currently registered to vote	36 (100)
Personal political activities in the last decade:
Voted in a national election	35 (97)
Contributed money to a political campaign	23 (64)
Attended a political campaign event	13 (36)
Volunteered time to a political campaign	7 (19)
Interactions with clinic patients in the last decade:
Had a discussion about voting	31 (86)
Initiated a discussion about voting	15 (42)
Discussed a politically oriented health care issue	30 (83)
Initiated a discussion about a politically oriented health care issue	17 (47)
Agree these politically oriented activities are appropriate in clinic:
Remind patients to vote through clinic brochures	30 (83)
Remind patients to vote through direct discussion	21 (58)
Inform patients about politically oriented health care issues through clinic brochures	18 (50)
Discuss politically oriented health care issues with patients	15 (42)

UPHS, University of Pennsylvania Health System.