Electronic medical record use and physician-patient communication: an observational study of Israeli primary care encounters.

Objectives: Within the context of medical care there is no greater reflection of the information revolution than the electronic medical record (EMR). Current estimates suggest that EMR use by Israeli physicians is now so high as to represent an almost fully immersed environment. This study examines the relationships between the extent of electronic medical record use and physician-patient communication within the context of Israeli primary care.

Methods: Based on videotapes of 3 Israeli primary care physicians and 30 of their patients, the extent of computer use was measured as number of seconds gazing at the computer screen and 3 levels of active keyboarding. Communication dynamics were analyzed through the application of a new Hebrew translation and adaptation of the Roter Interaction Analysis System (RIAS).

Results: Physicians spent close to one-quarter of visit time gazing at the computer screen, and in some cases as much as 42%; heavy keyboarding throughout the visit was evident in 24% of studied visits. Screen gaze and levels of keyboarding were both positively correlated with length of visit (r = .51, p < .001 and F(2,27) = 2.83, p < .08, respectively); however, keyboarding was inversely related to the amount of visit dialogue contributed by the physician (F(2,27) = 4.22, p < .02) or the patient (F(2,27) = 3.85, p < .05). Specific effects of screen gaze were inhibition of physician engagement in psychosocial question asking (r = -.39, p < .02) and emotional responsiveness (r = -.30, p < .10), while keyboarding increased biomedical exchange, including more questions about therapeutic regimen (F(2,27) = 4.78, p < .02) and more patient education and counseling (F(2,27) = 10.38, p < .001), as well as increased patient disclosure of medical information to the physician (F(2,27) =3.40, p < .05). A summary score reflecting overall patient-centered communication during the visit was negatively correlated with both screen gaze and keyboarding (r = -.33, p < .08 and F(2,27) = 3.19, p < .06, respectively).

Discussion: The computer has become a 'party' in the visit that demanded a significant portion of visit time. Gazing at the monitor was inversely related to physician engagement in psychosocial questioning and emotional responsiveness and to patient limited socio-emotional and psychosocial exchange during the visit. Keyboarding activity was inversely related to both physician and patient contribution to the medical dialogue. Patients may regard physicians' engrossment in the tasks of computing as disinterested or disengaged. Increase in visit length associated with EMR use may be attributed to keyboarding and computer gazing.

Conclusions: This study suggests that the way in which physicians use computers in the examination room can negatively affect patient-centered practice by diminishing dialogue, particularly in the psychosocial and emotional realm. Screen gaze appears particularly disruptive to psychosocial inquiry and emotional responsiveness, suggesting that visual attentiveness to the monitor rather than eye contact with the patient may inhibit sensitive or full patient disclosure.

Practical Implications: We believe that training can help physicians optimize interpersonal and educationally effective use of the EMR. This training can assist physicians in overcoming the interpersonal distancing, both verbally and non-verbally, with which computer use is associated. Collaborative reading of the EMR can contribute to improved quality of care, enhance the decision-making process, and empower patients to participate in their own care.