Background and importance Computerised physician order entry (CPOE) systems can enhance medication safety, but their implementation often faces hurdles, frequently due to user resistance. Implementation success depends on users’ acceptance of the system. Therefore, it is important to know the users’ views to tailor the implementation process and future optimisation of the system.

Aim and objectives Our aim was to examine physicians’ and nurses’ expectations before and opinions after the implementation of a CPOE system.

Material and methods We set up a survey combining validated tools and implementation-oriented questions to compare expectations before with opinions after CPOE implementation. The latter questions addressed quality of care, traceability of clinical decisions, readability, availability of patient information, quality of documentation, and time spent on documentation. Responses were rated on a five-point Likert scale. Additionally, respondents were asked to rate the implementation on a scale from 0 (negative) to 100 (positive). The survey was distributed to seven orthopaedic wards 4 weeks before and 3 months after implementation and anonymously collected. For the analysis of Likert responses, the number of responses of the upper and lower two Likert points was added up. Surveys with a non-response rate of >25% (excluding demographic questions) were not analysed.

Results The return rate was 36% (N=72) before and 26% (N=53) after implementation. Of the total of 125 surveys (25 returned by physicians and 100 by nurses), 121 were included in the analyses. The majority of respondents (60%–87%) expected CPOE implementation to improve the questioned criteria, except for quality of patient care (40% expected an improvement) and time spent on documentation (76% predicted an increase). After implementation, respondents reported improvements in readability (84%) and availability of patient information (57%); 78% of respondents indicated that the amount of time spent on documentation had increased. All other criteria were reported as improved, unchanged or worsened in approximately equal proportions. When asked to rate the implementation, the mean score was 56.5 ±30.0 before and 56.9 ±33.2 after implementation (p=0.809).

Conclusion and relevance The fact that positive and negative attitudes remained unchanged suggests that the opinions of doubters and enthusiasts were not significantly affected by the implementation. It might be worthwhile to choose different implementation strategies for these two groups.

Conflict of interest No conflict of interest