Changes in end-user satisfaction with Computerized Provider Order Entry over time among nurses and providers in intensive care units
This is a review of Hoonakker, P.T., Carayon, P., Brown, R. L., Cartmill, R. S., Wetterneck, T.B, Walker, J. M. (2013) article, Changes in end-user satisfaction with Computerized Provider Order Entry over time among nurses and providers in intensive care units.
The subject matter of the article under investigation focuses on Computerized Provider Order Entry (CPOE) and CPOE user satisfaction. Attempts to implement new health information systems relying on CPOE have traditionally been met with criticism and a certain degree of resistance among physicians and nurses. The authors of the study assert that previous literature and research efforts have not examined changes in user attitudes towards CPOE and other health information systems over time.
The purpose of the study was to identify if users’ attitudes and satisfaction levels change towards CPOE systems over time.
The authors of the article employed a mixed method research strategy using a survey as the primary data collection instrument. The survey is comprised of 15 closed-ended questions and contains two additional open-ended questions. The closed-ended questions of the survey represent the quantitative element of a mixed method study whereas the open-ended questions constitute the qualitative component. The research team also incorporated a cross-sectional design element to measure nurse and physician responses at two different points in time, although the nurses and physicians in each group are not necessarily the same participants.
The Provider Order Entry User Satisfaction & Usage Survey (POESUS), a validated questionnaire, was distributed to nurses and ordering providers in four ICUs at 3 months (n=177) and 12 months (n=220) after CPOE implementation. Each item was answered based on a scale ranked from 1 (never) to 7 (always) and an overall scale from 0 (lowest) to 100 (highest) was created to measure user-satisfaction.
Results from the study indicate nurses may become more accepting and satisfied with CPOE systems after a significant period of time passes, whereas providers’ general satisfaction levels remained almost the same. The additional time afforded to nurses enables them to gain hands-on experience with CPOE systems and seemingly reduces frustrations associated with preliminary CPOE implementation phases. Both the physicians and nurses liked the system because of the ease in reading orders, making orders, and the range of patient data they have access to. The study identified how CPOE systems can accelerate ordering management processes and integrate Clinical Decision Support (CDS) to improve safety and quality metrics associated with ordering functions. Instead of evaluating CPOE user feedback based on a single interval, the research team collected feedback three months after CPOE system implementation and again 12 months after implementation. The voluntary survey response rate for the first interval was 47% followed by 68% for the second interval.
At 3 months post CPOE implementation, surveyed nurses (n=121) reported below the scale midpoint (mean=48.6) and providers (n=54) reported mean 57.8 overall user-satisfaction. At the 12 month post CPOE implementation, nurses (n=163) reported and increased satisfaction (mean=56.8); however, providers remained around the same (mean=57.3).
Nurses and providers only experienced between six and ten hours of training prior to the implementation of the CPOE system. Whether a new information system is rolled out in a healthcare, business, or educational work environment, logic suggests end users will demonstrate frustrations related to overhauls in work processes; a characteristic likely exacerbated by insufficient training and support. Another criticism of the study pertains to response rates, or lack thereof, at both intervals. It is plausible to suspect nurses and providers who chose to participate in the survey may be individuals who are particularly frustrated by the CPOE system. Conversely, those who opted not to participate may encompass nurses and providers who are pleased or content with the CPOE systems. Such low response rates when dealing with user-experience ratings raises questions about whether the survey results represent the population they purportedly exemplify. Although the data collection instrument is intrinsically valid and reliable, those factors do not eliminate the need for a representative sample. The authors also rightly admit the study’s cross-sectional design render the results less meaningful since it is impossible to identify how many participants in the first survey interval participated in the second survey interval. At best the study results suggest time, as an independent variable, is potentially correlated with improvements in user satisfaction levels connected to newly implemented CPOE systems.
- Hoonakker, P.T., Carayon, P. , Brown, R. L., Cartmill, R. S., Wetterneck, T.B, Walker, J. M. Changes in end-user satisfaction with Computerized Provider Order Entry over time among nurses and providers in intensive care units. Journal of the American Medical Informatics Association: JAMIA, 20(2), 252-259. doi:10.1136/amiajnl-2012-001114. http://www.ncbi.nlm.nih.gov/pubmed/23100129