Short communication
Physician overestimation of patient literacy: A potential source of health care disparities

https://doi.org/10.1016/j.pec.2006.10.007Get rights and content

Abstract

Objective

To investigate physician overestimation of patient literacy level in a primary care setting.

Methods

The study sample consisted of 12 non-academic primary care physicians and 100 patients from a U.S. Department of Veterans Affairs Hospital in Houston, Texas. Patient literacy level was measured on a 1–4 scale using the Rapid Estimate of Adult Literacy in Medicine (REALM). Physicians rated each patient's literacy level on a corresponding scale. Chi-square was used to test for association of patient race/ethnicity and gender with: (1) patient REALM level and (2) discrepancy between patient REALM level and physician rating of patient literacy level.

Results

Patient REALM level was not statistically significantly associated with patient race/ethnicity or gender. Physicians overestimated the REALM level for 54% of African American, 11% of white non-Hispanic, and 36% of other race/ethnicity patients (p < .01).

Conclusion

Physicians commonly overestimate patients’ literacy levels, and this apparently occurs more often with minority patients, and particularly with African Americans, than with white non-Hispanic patients. This discordance in estimation of patient's literacy level may be a source of disparities in health care.

Introduction

Low literacy is widely known to be a major problem in today's health care environment [1], [2]. A growing number of studies link low patient literacy with poor outcomes [3], [4], including higher rates of hospitalization [5]. Patients with low literacy not only have difficulty reading printed health materials, they also struggle to understand technical terms, jargon, and complex concepts that are often embedded in physicians’ communication during the medical encounter [6]. For example, diabetic patients with low literacy misunderstood the content of the medical interview to a greater extent and had poorer recall of medical information than patients with higher literacy levels [7].

An important root cause of misunderstandings between patients and physicians may be physicians’ difficulty in detecting patients with low literacy. For many patients, the stigma of low literacy may lead to compensatory behaviors that make the patient's literacy difficult to characterize [8]. In addition, physical characteristics, such as appearance and speaking ability, are often inaccurate indicators of a patient's literacy level [4]. In such a context, physicians often make instantaneous, subconscious, and inaccurate judgments about what their patients can understand [9], [10]. For example, a study of resident physicians showed incorrect assessments of patient literacy levels in 34% of cases, and 95% of these assessments were overestimates [9]. This has important implications for education and counseling activities during the medical encounter, because overestimation may lead the physician to calibrate his or her language to a level that is not easily understood by the patient. In this study, we aimed to build on earlier work [9], [11] by examining experienced, practicing physicians’ estimates of patients’ literacy levels, and comparing these estimates to patients’ actual levels. In addition, we examined rates of physician over- and underestimation of patient literacy levels across patient race/ethnicity, because low literacy is known to disproportionately affect racial and ethnic minorities [2], and across pre-existence of a physician–patient relationship (i.e., “first visit” versus “repeat visit”).

Section snippets

Methods

Twelve full-time primary care physicians and 100 patients (4–14 patients per physician) were recruited at the Michael E. DeBakey Veterans Affairs Medical Center in Houston, Texas from June to October 2004. Participation was voluntary and written informed consent was obtained in advance from all participants. Our study protocol was approved by the Baylor College of Medicine institutional review board. In contrast to prior studies [9], [11], we recruited non-academic clinic physicians only.

Results

Of the 12 physicians we recruited, 2 were female. Physicians’ mean time in practice was 15.2 years (S.D. = 6.0). We approached 131 patients; 21 refused to participate, 2 were excluded due to confirmed acute visual impairment, and 8 failed to complete the protocol, resulting in a final sample of 100 patients. The mean age of these patients was 61.2 (S.D. = 11.2). Table 2 details the race/ethnicity and gender of the patients as well as the proportion reporting a pre-existing relationship with their

Discussion

Using a unique patient population and full-time, non-academic physicians averaging more than 15 years in practice instead of residents, we found corroborating evidence of the same physician overestimation problem identified previously [9], [11]. In 40% of our sample of U.S. veteran patients, physicians misjudged their patients’ literacy levels, and overestimates outnumbered underestimates nearly 2-to-1. We also found evidence that overestimation occurred more frequently with racial/ethnic

Acknowledgments

Support for this research was provided by the Health Services Research and Development Service of the U.S. Department of Veterans Affairs. Dr. Kelly is supported by career development award from the U.S. Department of Veterans Affairs. An earlier version of this manuscript was presented at the annual Research and Teaching Forum of the American Academy on Physician and Patient, Indianapolis, Indiana, October 2004. Our thanks to Leola Jones, BS, for data collection and assistance with analysis.

References (17)

There are more references available in the full text version of this article.

Cited by (252)

  • Spine Surgeon Estimation of Patient Health Literacy

    2024, Operative Techniques in Orthopaedics
View all citing articles on Scopus
View full text