Physicians' attitudes towards ePrescribing--evaluation of a Swedish full-scale implementation

Abstract

Background: The penetration rate of Electronic Health Record (EHR) systems in health care is increasing. However, many different EHR-systems are used with varying ePrescription designs and functionalities. The aim of the present study was to evaluate experienced ePrescribers' attitudes towards ePrescribing for suggesting improvements.

Methods: Physicians (n = 431) from seven out of the 21 Swedish health care regions, using one of the six most widely implemented EHR-systems with integrated electronic prescribing modules, were recruited from primary care centers and hospital clinics of internal medicine, orthopaedics and surgery. The physicians received a web survey that comprised eight questions on background data and 19 items covering attitudes towards ePrescribing. Forty-two percent (n = 199) of the physicians answered the questionnaire; 90% (n = 180) of the respondents met the inclusion criteria and were included in the final analysis.

Results: A majority of the respondents regarded their EHR-system easy to use in general (81%), and for the prescribing of drugs (88%). Most respondents believed they were able to provide the patients better service by ePrescribing (92%), and regarded ePrescriptions to be time saving (91%) and to be safer (83%), compared to handwritten prescriptions. Some of the most frequently reported weaknesses were: not clearly displayed price of drugs (43%), complicated drug choice (21%), and the perception that it was possible to handle more than one patient at a time when ePrescribing (13%). Moreover, 62% reported a lack of receipt from the pharmacy after successful transmission of an ePrescription. Although a majority (73%) of the physicians reported that they were always or often checking the ePrescription a last time before transmitting, 25% declared that they were seldom or never doing a last check. The respondents suggested a number of improvements, among others, to simplify the drug choice and the cancellation of ePrescriptions.

Conclusion: The Swedish physicians in the group studied were generally satisfied with their specific EHR-system and with ePrescribing as such. However, identified weaknesses warrant improvements of the EHR-systems as well as of their implementation in the individual health care organisation.

Figure 1

Number of ePrescriptions in Sweden per year, between 1983 and 2007. The first ePrescription was launched in 1983. After a new national strategy was decided at the end of the 1990's, the number of ePrescriptions in Sweden has escalated. Figure reproduced with permission [7].

Figure 2

Flowchart describing how ePrescriptions in Sweden are transmitted, stored and dispensed. The patient contacts the prescriber because of a medical need. The prescriber transmits an ePrescription to the online prescription repository. The patient may then request the ePrescription by contacting a pharmacist at any pharmacy in Sweden or a pharmacist at a mail-order pharmacy. When dispensing, the pharmacist collects information about the requested ePrescription from the online prescription repository and may also store dispensing information. Patients can reach the online prescription repository directly, through the Internet, and get information about his/her ePrescriptions.

Figure 3

Overview of the survey process of physicians.

Figure 4

Respondents' procedures when cancelling prescriptions. Respondents' answers to the question "Describe your actions when/if you want to cancel an ePrescription after the prescription is transmitted to the pharmacy." The respondents (%) on the y-axis and number of respondents (n) on top of the columns. The completeness rate was 0.89 (160/180).

Figure 5

Respondents' procedures when the ePrescription transfer is disrupted. Respondents' answers to the question "How do you proceed if/when you get a message that the ePrescription transmitting process is down, that is, it is not possible to transmit ePrescriptions? The patient is with you and is not in acute need of the drug". The respondents (%) on the y-axis and number of answers (n) on top of the columns. One respondent could give more than one answer. The completeness rate was 1.00 (180/180).