This paper described the organisation of a prototype preclinical patient-oriented educational scenario about PHRs in a medical informatics course. Students displayed, in general, satisfaction about the PHR they used and they were optimistic about using it in the future. They have valued most PHR’s main role as a medical record, but they have expressed some concerns about the PHR use by patients as a mere result of the opinion of the responsible physician towards its use. The simple prediction of their future attitudes towards PHR integration was found to be dependent on the level of PHR satisfaction that they gained through the educational encounter and experience gained, thereby indicating the pivotal role of good educational experiences and contemporary topic lab designs may play in improving PHR perspectives.
One can easily observe that the three features of Google Health which were rated higher were those which were assessed during the exercise part of the scenario. The fact that only some (and not all) features were part of the exercise may have led students to a slight overestimation of their significance with respect to the rest. The importance of assessing students in such an educational scenario is evident in this case. Assessment can be really considered as a tool to further improve the experience of students with the functions of a PHR in general, and was considered necessary as it may provide further means to emphasise the importance of using PHRs in clinical practice. Speaking about the importance of the Drug Interactions feature in PHRs, students seemed to like the way that it makes the patients more active when it is used (one should recall that the “patient-centric” scenario presented students with the chance to experience it as patients). Furthermore, as expected, the satisfaction concerning the Search for a Doctor feature was lower compared to other features, as this one is not properly functional internationally in the Google Health environment (the course was offered in Greek). Thus, the results of the searches were not as accurate when students searched for physicians outside the USA. Although it’s a critical feature for PHRs, as it connects patients to their physicians (PHRs to EHRs in general), the Sharing feature in Google Health corresponded to a relatively large percent (about 49%) of minimum-medium responses. A possible reason is that students found the presentation of this feature uninteresting, as the data shared were the same that they had to complete when training for the Profile Options feature. The statistically significant lower mean of dentistry students on their satisfaction of the Health Topics feature is probably due to the absence of presenting highly specific information on dental issues within the overall protocol.
The fact that the Physician’s Opinion factor concerns students the most in using Google Health, is probably due to their identity as future physicians. It also underlines the importance of having physicians with positive attitudes towards PHRs to facilitate PHR diffusion. It can also be noticed that Accessibility to the PHR (ability to operate a computer, familiarity with the Internet) is a factor which is particularly important to the use of PHRs. This is consistent with other results describing the reduced contact with PHRs that people with little computer skills have in relation to others . However, the fact that it was ranked lower than the other two seems to correspond to the increased familiarity of the new generation of health professionals (first-years) with computers and the Internet , as they pretended themselves to be the patients in the scenario herein. Gathering Anonymous Statistical Data from patients seems to be a major concern for them also, which presents a conflict with the Observing Health Trends aspect of the future utility of PHRs (they presented almost the same high to maximum answer percentages, around 65%). This conflict cannot, of course, be attributed to possible threats to a patient’s privacy, as the data collection was defined as anonymous. One plausible cause is that the students are concerned about the use of these data mostly for commercial profit, with less emphasis on healthcare improvements or scientific empowerment. The fact that the question referred to Google Health may be crucial here, because it may have directed student concerns to the importance of these reasons.
With respect to the likely future use of PHRs by patients, seemingly all areas described in the questionnaire provide statistical evidence about the future clinical practice of students; this is revealed by both the data and the students’ (qualitative) comments. The utility on the Up-To-Date Medical Record was rated high compared to all other aspects, which is consistent with the high rank that the Profile Options feature achieved. This observation is also consistent with one of the key advantages that can be derived from an interconnection between PHRs and the physician’s EHR system. The PHR utility in Physician-Patient Communication was rated lower than anything else, which outlines one of the major problems described about physicians and PHRs: change in the traditional relationship between them and the patients. It’s evident even in first-year students , that the physician’s role is such that it only supports communication in a certain way, so transferring this link into the Web through PHRs may be perceived as a danger of eliminating it.
The positive dependence between the two indices partially verifies the ideas about the usefulness of the undergraduate education of health professionals in the diffusion of PHR to the community of patients, although the experience that students gained was through a patient-oriented medical scenario. Achieving proper experience in such a scenario may improve the attitude of students towards PHRs, reflected by the FDI, so when they encounter them later in their education or clinical practice, they will be more urged to use them, as they will be suitably aware of the subject. The difference of EI-FDI dependence between the two student groups can be due to the fact that the scenario was mainly designed for medical students, as the topics discussed concerned mostly them. Although the simple linear model may not be the best fit for our dataset, it can provide useful insights on the scenario. Firstly, if EI = 0 was to be interpreted as no contact with the scenario, the level of FDI would be 0.22, which reflects the estimate of the PHR integration (approximately to a level of 20% of the total patients) without the existence of the scenario. Secondly, a student’s maximum satisfaction (EI = 1) corresponds to FDI = 0.914, which leads us to another conclusion: the education-attributed integration is 0.694, which corresponds to a potential 0-70% increase of the patients that will be affected by the physician’s positive attitude towards PHRs.
Finally, the current piece of work is by no means moderated by Google’s recent announcement their retirement from the PHR Web service (note that it was reported that Google Health “is not having the broad impact that we hoped it would” ). In fact, this underlines the emerging importance of physician education on PHRs, as having more PHR-educated physicians would be crucial in creating large-scale studies about the features that are most valued by them in a PHR product. This would attract major players in Web market to create refined and easy-to-use PHR products that emphasize those features.
Since no stratification of students was conducted with respect to their PHR background or, generally, electronic record keeping familiarity, this might introduce some bias in the answers of the questionnaire, as our sample might have displayed generally better (or worse) attitude towards PHRs relative to a sample that had no members with previous PHR interaction. Another factor that introduces positive bias to the answers of our sample is the fact that the questionnaire was completed inside the educational activity, as students might have considered it as a part of a common lesson evaluation.
The scenario through which the EI and FDI values were acquired was patient-oriented. The experience that medical students will get through a PHR scenario may be one of a different educational perspective, so one cannot really generalise the dependence in their values well. The fact that the students were first-years presents some problems when it comes to the part of the questionnaire related to their future expectations, as their judgement on that is not based on real experience, but mere hypothetical thinking. As a result, FDI, which is directly related to these questions, suffers from interpretation bias, as it is just a measure of the estimate of students’ future attitude. The actual level of the PHR integration probability may differ, if we were to study the long-term students’ behaviour, after they had obtained enough clinical experience (e.g. during the last two years of their studies). These facts may undermine any possible practical importance that the results would have on further applications.
Implications for Future Research
It is suggested that medical informatics courses that require sufficient medical background, such as those on EHRs and PHRs, should be placed towards the clinical part of the medical curriculum, to facilitate the use of the items learned . Preclinical introductory courses about these items can also exist to provide information literacy and help students develop practical skills . Considering all these, educational activities based on PHRs should be formally described, so that medical schools can easily integrate them in their curricula. Interventional studies comparing “educated” groups against “uneducated” groups of students could also be organised, so as to fully determine if the presence of an educational scenario were to improve student attitudes towards PHRs.
Studying the results of a similar physician-oriented educational process (teaching different skills to students, such as interconnecting an EHR system with PHRs) would make apparent if undergraduate scenarios in PHRs have a direct effect to better future PHR attitudes. A comparison between the results of the two educational approaches afore-described, i.e. patient or physician-oriented, would also be made possible in that case. It can be stated that, besides the serious limitations that go with them, the first years’ results may be used in a later study concerning the temporal evolution of an intervention’s effect on the sample (by us or by others who want to address the same subject).
Finally, there are obviously more PHR products which include all the characteristics needed in a similar educational approach (Indivo [59, 60], Microsoft Health Vault ). The results from the application of a similar scenario, which would use a different PHR, could be compared to the present ones. Moreover, an educational scenario that would incorporate data from multiple PHRs could be created, thereby making possible the study of the effect of PHR product on the students’ opinions.
In a “self-caring society”, patients would have liked to take full part in deciding about their treatment in a symmetric and negotiated relationship with healthcare professionals; for the later to be able to react on their patients’ empowerment they need to be educated and familiar with PHR systems that patients are likely to interact with. For example, patients joining the PatientsLikeMe community go online to (not only discuss health and daily living) but to share detailed health data. Strictly speaking, of course, PatientsLikeMe is not a PHR system, but rather a “shared” online platform where patients “share structured information about symptoms, treatments, and outcomes, view individual and aggregated reports of these data, and discuss health and garner support on forums and through private messages”. Members of PatientsLikeMe offer one another support based on their own personal experience and advise each other on both medical issues and how to improve day-to-day life .
Recent studies with such PHR resembling systems, provide evidence that patient-reported data and outcomes, offer a unique real-time approach to understand utilization and performance of treatments across many conditions and potentially identify targets for treatments .
But since PatientsLikeMe is something between a classical PHR and an information searching platform, one could indirectly assume that the effects web searching and information sharing technologies bring about with respect to the patient empowerment and the change of doctor-patient relationships [64–66] are related to our study. This makes stronger the need to keep a record of how professionals are educated with respect to these systems and how this education could be potentially modeled to enable future optimization of the societal benefits outweighing potential technology threats.