What Do Doctors Say When Prescribing Medications?: An Examination of Medical Recommendations From a Communication Perspective

Abstract

We described physician usage of persuasive strategies pertaining to four dimensions of medical recommendations given during naturally occurring clinical visits—problem seriousness, treatment effectiveness, patient's self-efficacy, and potential limitations with the recommended treatment. We then examined the impact of these strategies on patient satisfaction and intention to follow physicians' medical advice. An analysis was conducted of 187 transcripts of audio-recorded outpatient visits during which a new medication was prescribed, augmented with patient and physician surveys. Two-hundred forty-two cases of new medication prescription were identified, and each case was coded into categories describing physicians' prescription-giving behaviors. In most cases, physicians addressed only one or two of the four dimensions of medical recommendations when they were prescribing new medications to their patients. In about one-third of visits, none of the four dimensions was addressed. However, physician use of persuasive strategies pertaining to the four dimensions did not appear to have any significant impact on patients' satisfaction with the visit or intention to follow their doctor's advice. The implications of the findings are discussed in light of the study's limitations and directions for future research.

ACKNOWLEDGMENTS

The authors acknowledge the coding assistance provided by Kristen Gottschalk and Vida Z. Gabe. This publication was made possible by grant UL1 RR024146 from the National Center for Research Resources (NCRR), a component of the National Institutes of Health (NIH), and NIH Roadmap for Medical Research. Its contents are solely the responsibility of the authors and do not necessarily represent the official view of NCRR or NIH.

Notes

¹Since the two dependent variables of this study (patient satisfaction with visit and intention to follow physician advice) were both measured at the visit/patient level rather than the medication level, it was necessary to create an index of IMMA usage for each visit.

²STATA offers a very large range of programs for analysis of complex sample survey data. In this study, the svyset commands were used to specify the stratification scheme, sampling weights and primary sampling units for the data. The svy commands were used to produce estimates that were corrected for the complex design of the survey.

³The distribution of satisfaction with visit was negatively skewed (skewness = −1.28, kurtosis = 1.37); the distribution of intention to follow advice was positively skewed (skewness = 2.81, kurtosis = 10.39); the distribution of trust was positively skewed (skewness = 1.83, kurtosis = 3.98).

⁴Satisfaction was log transformed and both intention to follow advice and trust were inverse transformed.

⁵The number of dimensions addressed was positively related to patient's age, b = .02, t(40) = 3.47, p < .001, and negatively related to number of prior visits with patient, b = −.22, t(40) = −2.6, p < .05.

⁶Details of these analyses are available from the first author.

⁷Patient trust in physician was strongly associated with satisfaction with visit, b = −.55, t(40) = −7.68, p < .001, and intention to follow doctor's advice, b = .53, t(40) = 7.15, p < .001.