Discrete Choice Experiments (DCEs) are increasingly used in studies in healthcare research but there is still little empirical evidence for the predictive value of these hypothetical situations in similar real life circumstances. The aim of this paper is to compare the stated preferences in a DCE and the accompanying questionnaire with the revealed preferences of young parents who have to decide whether to vaccinate their new born child against hepatitis B. A DCE asking parents to decide in which scenario they would be more inclined to vaccinate their child against hepatitis B. The stated preference was estimated by comparing the per respondent utility of the most realistic scenario in which parents could choose to vaccinate their child against hepatitis B, with the utility of the opt-out, based on the mixed logit model from the DCE. This stated preference was compared with the actual behaviour of the parents concerning the vaccination of their new born child. In 80% of the respondents the stated and revealed preferences corresponded. The positive predictive value is 85% but the negative predictive value is 26%. The predictive value of the DCE in this study is satisfactory for predicting the positive choice but not for predicting the negative choice. However, the behaviour in this study is exceptional in the sense that most people chose to vaccinate. Future studies should focus on behaviours with a larger variance in the population.

As the implementation of Electronic Medical Records (EMRs) in hospitals may be challenged by different responses of different user groups, this paper examines the differences between doctors and nurses in their response to the implementation and use of EMRs in their hospital and how this affects the perceived quality of the data in EMRs. Questionnaire data of 402 doctors and 512 nurses who had experience with the implementation and the use of EMRs in hospitals was analysed with Multi group Structural equation modelling (SEM). The models included measures of organisational factors, results of the implementation (ease of use and alignment of EMR with daily routine), perceived added value, timeliness of use and perceived quality of patient data. Doctors and nurses differ in their response to the organisational factors (support of IT, HR and administrative departments) considering the success of the implementation. Nurses respond to culture while doctors do not. Doctors and nurses agree that an EMR that is easier to work with and better aligned with their work has more added value, but for the doctors this is more pronounced. The doctors and nurses perceive that the quality of the patient data is better when EMRs are easier to use and better aligned with their daily routine. The result of the implementation, in terms of ease of use and alignment with work, seems to affect the perceived quality of patient data more strongly than timeliness of entering patient data. Doctors and nurses value bottom-up communication and support of the IT department for the result of the implementation, and nurses respond to an open and innovative organisational culture.

Background: We developed a system to automatically classify stance towards vaccination in Twitter messages, with a focus on messages with a negative stance. Such a system makes it possible to monitor the ongoing stream of messages on social media, offering actionable insights into public hesitance with respect to vaccination. At the moment, such monitoring is done by means of regular sentiment analysis with a poor performance on detecting negative stance towards vaccination. For Dutch Twitter messages that mention vaccination-related key terms, we annotated their stance and feeling in relation to vaccination (provided that they referred to this topic). Subsequently, we used these coded data to train and test different machine learning set-ups. With the aim to best identify messages with a negative stance towards vaccination, we compared set-ups at an increasing dataset size and decreasing reliability, at an increasing number of categories to distinguish, and with different classification algorithms.Results: We found that Support Vector Machines trained on a combination of strictly and laxly labeled data with a more fine-grained labeling yielded the best result, at an F1-score of 0.36 and an Area under the ROC curve of 0.66, considerably outperforming the currently used sentiment analysis that yielded an F1-score of 0.25 and an Area under the ROC curve of 0.57. We also show that the recall of our system could be optimized to 0.60 at little loss of precision.Conclusion: The outcomes of our study indicate that stance prediction by a computerized system only is a challenging task. Nonetheless, the model showed sufficient recall on identifying negative tweets so as to reduce the manual effort of reviewing messages. Our analysis of the data and behavior of our system suggests that an approach is needed in which the use of a larger training dataset is combined with a setting in which a human-in-the-loop provides the system with feedback on its predictions.

Currently electronic medical records (EMRs) are implemented in hospitals, because of expected benefits for quality and safety of care. However the implementation processes are not unproblematic and are slower than needed. Many of the barriers and facilitators of the adoption of EMRs are identified, but the relative importance of these factors is still undetermined. This paper quantifies the relative importance of known barriers and facilitators of EMR, experienced by the users (i.e., nurses and physicians in hospitals).