Effects of online palliative care training on knowledge, attitude and satisfaction of primary care physicians

Palliative care (PC) in Spain can be delivered to patients at hospital and in the last months of life mainly at home, by primary care physicians (PCPs) supported by specialist PC teams, not equally distributed all over the country, and with access to an inpatient facility when required. The Spanish PC Strategy recommends an intermediate level of training for PCPs in order to provide them with knowledge and skills. Most of the training consists of face-to-face courses, but increasing pressures on PCPs have resulted in fewer opportunities for provision and attendance to this type of training.

Training needs of PCPs in PC have been described at great length in scientific literature [1‐4]. A systematic review of educational methods in PC for PCPs showed that those educational activities with multi-faceted methodology offered better results than those using simple didactic approaches [5]. The only study using Internet as supporting element, together with a joint approach, showed significant outcomes in terms of palliative treatment choice, and regarding PCPs' attitude and satisfaction in the intervention group.

E-learning or learning facilitated and supported through the use of information and communication technologies offers a learner centred model consistent with the adult learning theory [6, 7], where a direct and active learner involvement is required in order to obtain a subsequent behavior change. E-learning puts learners in control of their own learning in comparison with traditional instructor centred teaching. On-line educational programs offer practical benefits for learners who work well with computers and have access to Internet. It provides a real alternative to traditional methods of education because of its flexibility and interaction capabilities and is rapidly gaining in importance [8]. The efficacy of on-line continuing medical education is already quite well established in two aspects: knowledge acquisition, with results comparable to other traditional models, and learner satisfaction, with higher satisfaction results compared to other models. However, there is uncertainty in terms of potential benefits of asynchronous communication and interaction among participants, and the effect of the learning process. Asynchronous communication, or the relay of information with a time lag, e.g. in discussion forums and e-mails, provides benefits for learning. Some proposed benefits are: that it helps learners to internalize and process information, it offers all learners equal opportunities to participate, it needs a lower hardware and network specification and it is very flexible. The main disadvantages to asynchronous communication are the waste of time while waiting for a response, and the learner's isolation [9].

There has been less research in terms of the effectiveness of on-line continuing medical education, in order to identify whether said model has an impact on clinical practice; and those few known results are mostly based on the physicians' own perception, instead of objective clinical measurements; therefore, these outcomes are still uncertain [10‐13]. Data from a systematic review [14] and from other studies [15] have detected discreet improvements of 4%-5%, or none at all. In the topic of PC and cancer the on-line model has also been successfully used for end-of-life care consultation [16], for pain management [17], or to obtain clinical information [18]. However, the effectiveness of on-line PC training for PCPs is currently uncertain, regarding improvement in patient's quality of life and main caregiver's satisfaction, which points out to the need of a clinical impact assessment through objective measurements [5, 19].

On-line continuing medical education is promising in terms of its ability to improve clinical practice, as long as the information provided, dissemination planning and pedagogical approach are accessible, useful, high quality, and focused on the learner [12, 20]. To determine its effectiveness in PC training for PCPs is important due to the fact of increasing constrains of attendance to face to face training and the quickly development of new technologies applied to instruction.

Planning and design of on-line training is particularly complex, as it involves multidisciplinary teams working together in order to shape educational materials according to the chosen pedagogical approach. Didactic resources must be organized and scheduled, tutor-mediated interaction between learners and teaching staff must be planned, and satisfaction and efficacy in terms of learning must be provided.

The authors have developed an on-line educational model for PC which has been applied to PCPs to measure its effectiveness regarding knowledge, attitude towards PC, perception of confidence of symptom management and communication with advanced cancer patients requiring PC and their relatives, and PCPs' satisfaction, in comparison with a control group and random allocation.

The effectiveness evaluation at 18 months, through objective clinical measures, of the impact on the quality of life of patients managed by the physicians, and the main caregiver's satisfaction will be addressed in a different paper.

Experience Setting: the training model being assessed has been applied during 2009 to PCPs from primary care centers in Spain. Participants were recruited by informing specific groups about the project, and offering teaching credits, participation as collaborator in the final publication of patients' results and financial incentives (80 euros per patient included). Participants had to be: PCPs from the Spanish NHS (with a practice of their own, substitute position ≥ two years, acceptable level of English in order to read medical texts, and Internet access). Those PCPs included were randomized (the physician is the unit of randomization) in two groups by computer-generated exchangeable sequence of two-blocks kept in opaque envelopes. Allocation was conducted when PCPs agreed to take part, by a researcher who didn't know the random sequence, and a sequential numerical code was then assigned. The intervention group had access to the on-line program for PC self-training, while the control group did not, but could voluntarily receive or not the usual PC training offered for his working area (traditional training).

Educational Contents: it was prepared by the authors according to clinical competences of PCPs in PC, as recently defined in the specialty program [21], and distributed in four modules (Figure 1). These modules were divided into subjects including: objectives, contents directed to clinical practice, with clear and concise explanations, PC bibliography and websites. These websites were chosen according to a previous pilot study, in which 27 PCPs assigned a score to the best websites based on their clinical usefulness, the quality of their information, and their accessibility. Each subject included links to explanations for the presentation, and self-guided questions which could be reviewed several times, without a final score.

Technological Platform: E-Learning platforms, also called VLE (Virtual Learning Environments) or LMS (Learning Management Systems), allow access to instructional information from any location and any time by using a personalized identification, making its use compatible with other activities. At the moment, there is no clear differential concept of technological platforms that facilitate learning and the variety of denominations are defined in terms of their functionality or a description of their functions and tools, instead of a definition consistent with the final objectives of the platform. All these platforms use a system or combination of rules, process or workflows that effectively manages any type of digital information, whether text, images, video, documents, sound files, etcetera. The main concept behind any Content Management System is to make the digital information available for inter-office or on-line edition.

Moodle is a very popular e-learning platform, which is widely used in health education contexts because it allows instructors to prepare and store educational resources easily, in order to deliver them to potential users. It was selected for our study in order to provide physicians with a common repository for the educational resources used in the current experience [22‐24]. The Dreamweaver tool was used for content design, and the platform resources (Figure 2) were prepared based on Word documents. Evaluation forums and questionnaires were conducted, complemented by other interactive resources (images, diagrams, videos, interactive webpages, among others).

Assessment: The quality of on-line training process, knowledge, and attitude towards PC were assessed, as well as satisfaction with the teaching activity.

Variables registered on-line were:

The study involved 169 PCPs throughout the whole country, with 85 in the intervention group and 84 in the control group. Five participants (three from the intervention group, two from the control group) were excluded for not providing data; they were three men and two women, all of them on payroll except one substitute. Finally 82 participants per group were analyzed. Baseline characteristics (Tables 3, 4) were balanced with a minimal difference in terms of previous PC training in the intervention group (78%) vs. the control group (70%). Most participants were family doctors (through Medical Residency Training), women, age > 40 years, wide professional experience, on payroll, with recent PC training, and most of them seeing two-four patients per year requiring PC. Most participants had support by the area PC Team, and its performance was considered from good to very good; their perception of confidence in symptom management was acceptable-good, as well as confidence in communication; they agreed-highly agreed that PC for advanced cancer patients should be part of care offered by PCPs.

The outcomes of the first stage of this study have shown that, after an on-line PC training program targeted to PCPs, using a Moodle Platform, with tutoring, and developed during 75 days, there was an increase of knowledge of 14%-20%, there was a significant increase in the perception of confidence in symptom management and communication, and overall satisfaction was good-very good for most participants. The cost for the education consisted of time spent by the authors in the implementation of the program and logistics. We value the time spent as very positive if we consider that the education model can be offered to other physicians interested in PC education in the near future.

These outcomes are comparable to other studies for on-line educational intervention, such as those reported about pain management and other oncology symptoms, or management of last hours of life [8‐12, 26‐29]. Most participants assigned a score of "good-very good" to the quality and usefulness of educational contents, with recommendation to other colleagues to take part, and agreement that the integration of platform resources facilitated the learning process. Even though some technical aspects to be improved were pointed out, the overall difficulty in terms of using the platform was considered low.

The decrease in knowledge in the control group, and specifically in the sub-group of 11 who received some kind of education, could be explained by the fact that the majority did not report any educational activity, and only four participants of 11 attended courses with more extensive subjects, while the rest reported self-study, perhaps with more emphasis on specific subjects, not so wide as those explored by the knowledge test. Regression to the mean could be another explanation. The probability of chance is low.

Confidence in communication increased significantly in one to three categories in the intervention group, with a marginal significance in the comparison among groups and low effect size in the data imputated. The subject of communication with the patient and the caregiver was elaborated through visual sketches and specific sentences that, in spite of being short in extension, could have made an impact on the confidence of the participants.

These results must be interpreted within the context of PCPs participating in this study with wide professional experience, knowledge, confidence in terms of symptom management and patient communication in PC from acceptable to high, frequently supported by the local PC team, and with a very high opinion about the integration of PC within Primary Care. Results could not necessarily be reproduced in another PCP profile, particularly in those with a low opinion about the integration of PC within primary care, and who are probably not able to read medical texts in English.

The variety of opinions by PCPs in terms of the on-line design of that training provided might be associated with different learning styles. Some studies about on-line training for healthcare professionals [30] suggest that those with an analytic learning style tend to respond better to less structured, long Web-pages, with an in-depth instead of a wide approach; on the other hand, those professionals with a more holistic style may respond better to less structured educational materials, shorter web-pages, and with a wider instead of an in-depth approach. It has been suggested that on-line education could be significantly encouraged by adapting teaching methods to analytic-holistic learning styles; this has not been thoroughly investigated in terms of medical education.

Another aspect about which there is limited knowledge is the emotional impact of terminal patient care, and how PCPs manage the relationship with patients and their family. Although this aspect was included in module 4 in our study, no specific questions came about from any participants about this issue. The study of this specific topic through a qualitative methodology, as has been carried out in other related topics [31], would allow us to include a future on-line module regarding emotional needs of PCPs, and tackle the issue at the same time.

Education using e-learning platforms has a highly positive and consistent effect when compared with no educational intervention, and seems to have the same efficacy and effectiveness than the traditional training system. A meta-analysis comparing medical knowledge acquired through internet vs. no intervention reported a considerable benefit in favor of Internet [32]. Even though these outcomes might be affected due to confusing factors or chance, it has been suggested that learning effectiveness should be currently investigated comparing different Internet interventions, instead of Internet vs. no intervention.

On-line PC training is increasingly spreading [33], and the on-going studies [34‐36] will help to clarify its effectiveness. In our study, on-line training has been effective in improving PCPs' knowledge and attitude, and therefore we consider it a useful training tool.

PC training for PCPs using an on-line educational model with a Moodle Platform and tutorship, showed an increase of knowledge of 14%-20%, and a significant increase in the perception of confidence in symptom management and communication, in comparison with a control group that received traditional methods of education or no educational activity at all. PCPs satisfaction was high.

According with the results this on-line educational model seems a useful tool for PC training in PCPs who have a high opinion that PC should be integrated in primary care, able to read English medical texts and supported at large by the PC team. Some limitations signaled are related with the current situation of medical practice.

The results of this study support the suggestion that learning effectiveness should be currently investigated comparing different Internet interventions, instead of Internet vs. no intervention.

The second phase of this study will examine on-line training effectiveness at 18 months, and its impact upon patient's quality of life and caregiver's satisfaction.