Prior to the implementation of Electronic Medical Record (EMR) web access for patients at a large integrated delivery systems, we surveyed physicians’ attitudes. Our web based questionnaire revealed largely positive attitudes about access. The exceptions included abnormal reports, progress notes, and e-care. A factor analysis identified the group of physicians who didn’t view patients as partners felt most negative about the process.

Although patients have had legal access to their medical records for many years, several institutional, provider-specific and logistical barriers made this difficult. In response to various pressures (including HIPAA), institutions worldwide are making this access easier by securely providing some information or even two-way communication electronically. Many initial reports focused on the uncharted risks and drawbacks of this access.¹

We administered a web-based questionnaire to 150 email-using physicians; non-responders had follow-up by phone or email. The questionnaire was developed through consensus, using a Likert scale and including a patient as partner subscale. Analysis was done via Wilcoxon rank-sum, principle components factor analysis, and a multivariate linear model.

91 (61%) of physicians responded; 78% of the non-responders were contacted. Figure 1 reveals the general results. Most felt the medication list, normal studies, prescription refills, appointments, and referrals should be provided to the patients (Wilcoxon, all p<.05). However, they felt progress notes, abnormal labs, and care over the internet should not be provided. They were more positive than negative about the effects of the system, (32–46% agree; 30–40% neutral) and 66% would not give any care over the internet. The factor analysis revealed a subset of ‘control’ physicians who did not see e-care as improving quality and did not want two-way messaging.

Despite previous comments, physicians not exposed to patient access to their own EMR saw many aspects of the policy as positive. Physicians who do not see patients as their partners are more likely to be negative; this group can be identified and their concerns addressed. Further work is being done to compare the after-implementation effects with these preliminary results.