Abstract
We conducted a systematic review of the perceptions of physicians about clinical supervision and educational support via videoconferencing. A search of literature databases, reference lists and specific journals was performed for relevant articles published between 1990 and 2013. A total of 1288 studies were identified, of which 13 fulfilled the inclusion criteria for review. Six studies concerned education, six concerned clinical practice and one concerned supervision. The studies employed a wide variety of methodologies, including quantitative and qualitative techniques, so a meta-analysis was not practicable. Overall satisfaction and acceptance rates were reported in nine studies, mainly using Likert scales. Several positive aspects of videoconferencing were reported, including increased education and clinical practice support, and autonomy for rural areas. The main negative aspects related to the technology itself, poorer interaction and decreased rapport building when using videoconferencing. There is a paucity of literature regarding the perceptions of videoconferencing-based supervision of junior doctors in rural areas. No studies have considered measures of cost effectiveness. More rigorous study design in future research is recommended.
Introduction
Improving primary health care for the rural population has been an important goal of the Australian government for the last two decades. A major challenge in rural communities is training and retention of medical staff. The most recent Health Workforce Australia document 2025 outlined significant concerns about predicted shortages of rural doctors and their unequal distribution between metropolitan and rural cities. 1 Doctors in rural communities often feel isolated and assume greater responsibility with a more varied caseload than their urban counterparts. 2
Most studies of the retention of physicians in rural areas have found that clinical supervision, education, training and professional support are necessary and vital.1,3 However, although direct specialist outreach and support visits from tertiary centre doctors are requested, they are often unattainable or unsustainable. Lack of academic support and isolation affects rural practice and the satisfaction with the service provided. 4 While face-to-face outreach specialist services attempt to reach rural areas and provide support and training to rural doctors, there have been problems with the variability, quality and irregularity of such methods. 4 Junior doctors are often forced to make decisions without supervision from more experienced clinicians, which can be extremely stressful.
Telemedicine has been trialled in rural areas with successful outcomes.5,6,7 Supervision of doctors using videoconferencing may reduce stress and improve decision making regarding health care needs for the rural population. 8 More widespread use of telehealth models for care and supervision has been recommended to relieve the pressure on rural doctors in Australia. 1
A review of the videoconferencing network in Queensland over a two-year period found that its main uses were education and professional clinical support (60%), and administrative support (40%). The professional development component included grand rounds, tutorials, meetings and working groups, all of which would have been unavailable without videoconferencing. 9 There were similar findings in Western Australia. 10 While there have been a number of studies reporting satisfaction levels, treatment outcomes, and costs for patients whose care is provided through videoconferencing, there are fewer studies which have focused on physicians’ perceptions using videoconferencing for training, education, and clinical supervision.11,12
Therefore the aim of the present study was to systematically review the literature on research which has addressed the question “What are the perceptions of physicians using videoconferencing for clinical supervision and educational support?”
Methods
A systematic review was conducted using the PRISMA guidelines and the Systematic Reviews in Healthcare methodology.13,14
Search
The search strategy was designed to identify relevant material published from 1990 to 2013 inclusive. An initial search of MEDLINE (OVID) and the cumulative index of nursing and allied health literature (CINAHL) was undertaken to identify suitable keywords in the title, abstract and subject descriptors. Medical Subject Headings (MESH) terms and synonyms used in various databases were included to perform an extensive search of the literature. Various search terms were applied in different databases in order to capture all relevant articles. The common search terms used were: physician*, “medical education”, or ‘rural or remote or regional’, “consultation remote” or “teleconsultation” or telemedicine* or telehealth, videoconferenc*, teach*, supervision, mentor*, and perception. See Appendix 1 for a full list of MESH terms and the search strategy.
Information sources
Studies were identified by searching electronic databases and scanning reference lists of key articles. Limits were applied to capture articles from the years 1990-2013, extracting only English language articles for review. The search was conducted in the following databases: CINAHL, MEDLINE (OVID), EBSCOhost, Informit, SCOPUS, PsycINFO, Cochrane Database of Systematic Reviews, Informit A+Education and ERIC. In addition, searches were conducted in the following journals: Journal of Telemedicine and Telecare, Rural and Remote Health, and Telemedicine Journal and eHealth. No searching of the grey literature, such as non-commercial published reports, theses, conference proceedings and working papers or unpublished material, was conducted.
Eligibility criteria and study selection
Studies were considered eligible for inclusion if they met the following criteria: (1) the participants were medical students or qualified physicians; (2) the intervention described was videoconference-based education, training or supervision using either IP or ISDN connections; (3) the outcomes focussed on perceptions, satisfaction and effectiveness of such methods; (4) the level of evidence according to the National Health and Medical Research Council (NHMRC) was greater than a case study. 15
Exclusion criteria were: (1) study participants from other health professions or most participants from other health professions. This included areas such as nursing, physiotherapy, occupational therapy, speech pathology, psychology, pharmacy and social work; (2) other interventions such as Internet-based online teaching and video-streaming studies; (3) case studies; (4) teaching for surgical procedures; (5) patient centred outcomes.
Abstraction and analysis
Articles were collected and reviewed in several stages. First, the titles and abstracts of articles were screened for eligibility and relevance. Other studies were recommended by experts in the field. Where articles did not provide enough information for initial selection, the full article was retrieved and reviewed for eligibility. Full article review and extraction was conducted, with the quality and methodological rigour of studies being evaluated according to NHMRC guidelines. 15 The studies were mainly qualitative in design. Formal meta-analysis was not possible, given the variation in methodology and high proportion of qualitative data. Articles were grouped in broad categories according to subject. These included education, clinical practice and supervision. Articles were then compared in relation to participants, study design and main findings. Any disagreements between the three reviewers were settled by consensus.
Results
A total of 1288 journal articles were obtained. Following removal of duplicates, 976 articles were screened by title and/or abstract. A further 927 articles were excluded having not met the inclusion criteria. The majority of these articles did not relate directly to telemedicine or videoconference-based studies. The remaining 49 articles were given full review, and 13 met the inclusion criteria. Of these 49, five were demographic studies, 15 included other health professions, five were patient based studies, four were non videoconference-based studies (e.g. they used still images or telephone consultations), six were studies that did not include perceptions/evaluations by participants, and one was a case study. After following all inclusion and exclusion criteria, 13 studies were included in the review (Figure 1).
Flow diagram for study selection process.
Characteristics
Study characteristics.
Comprehensive study characteristics.
VC=Videoconference, FTF=Face-to-face, CME=Continuing Medical Education, GP=General Practitioner, RN=Registered Nurse, NP=Nurse Practitioner, PA=Physician Assistant, PCP=Primary Care Physicians; YOE=Years of experience in medicine; ED=Emergency Department.
Videoconferencing
All studies used videoconferencing. ISDN transmission was used in nine studies (69%) and IP was used in one study (8%). Of the ISDN studies, six used 384 kbit/s (75%), one 256 kbit/s (13%), one 128 kbit/s (13%) and one 112 kbit/s (13%). Three articles did not specify configurations (23%). One study specified the use of a connection at 15–30 frames/s for videoconferencing (Table 1).
Study methodology
Nine studies (69%) used Likert or simple yes/no questions17–21,23–25,27 and six (66%) of these included a qualitative component i.e. (open ended questions or comments section).17–21,23 One study used a four-point scale survey which rated in an opposite fashion compared with the traditional Likert scale. 16 Only two studies (15%) used a purely qualitative component.26,28 Of these qualitative studies, findings were obtained through semi-structured interviews and observation field notes. One study used control participants. However, this was not matched in participant numbers. 16 Only one study focussed on pre and post outcomes, 21 while the rest focussed on post study outcomes.
Participants
Physician’s perceptions were examined exclusively in nine (69%) of the studies included in the review.17–19,21,23,24,26–28 In the remainder, participants included medical students, 16 managers, 22 nurses, nurse practitioners, physician assistants, psychologists and osteopaths.20,25
Ten articles (77%) described the medical specialities involved.17,19–22,24–28 The majority of medical specialities were General Practitioners or Primary Care Physicians in rural areas. Nine studies (69%) included information relating to rural or remote status of health centres.17,18,20–23,25–27 Of the 12 articles with physicians as participants, only three (25%) studies mentioned years of experience.17,21,28
Interventions
Six studies (46%) focussed on videoconferencing for continuing medical education or training (Table 1). One of these used control participants, comparing face-to-face versus videoconferencing lectures. 16 Clinical practice interventions were studied in six (46%), with one of these assessing medical education. 23 Only one (8%) study examined supervision in clinical practice. 28
Findings
Overall satisfaction and acceptance was determined, mainly through Likert scales in nine (69%) of the studies.16,18,19,21–25,27 The terms used to measure this were satisfaction, acceptance, impression, successfulness and usefulness. In the seven studies that used Likert scales to deduce this (1 = poor satisfaction, 5 = high satisfaction), the results ranged between 4 and 5, with a mean of 4.3. One study used a four-point scale to measure overall acceptance and satisfaction between face-to-face lectures and videoconference lectures amongst medical students. This scale used an opposite measure to the Likert scale in which 1 = excellent and 4 = poor. This study found a similar result to the other studies, the authors concluding that the high overall satisfaction with videoconferencing was because travel in rural communities was less attractive. However, there was slightly less satisfaction in the videoconference group. 16
Positive aspects of videoconferencing
Several positive aspects of videoconferencing were reported in the studies:
Educational support and supervision was measured in four studies.17,22,27,28 Perceptions included increased support for continuing medical education and increased supervision amongst professionals. Increased access to support was also mentioned in conjunction with decreased feelings of professional isolation.
20
Clinical practice support and knowledge was identified with high satisfaction ratings in eight studies.17,18,20,21,23–25,27 This included improvement in clinical skills training, and high relevance to clinical practice. Improving patient care was reported in six studies, with particular improvement in rural areas.18,20,22,24,25,27 Only one study commented on pre and post testing knowledge rising from 60% to 80% overall.
21
Organisational benefits of telemedicine were identified in two studies, specifying improvement in autonomy for rural centres.22,23 Improvement in communication and collegiality between institutions and providers was also mentioned.24,27 The benefits of decreasing or avoiding travel for physicians and patients were reported in five studies. This included higher levels of acceptance in areas which required further to travel and was increased in communities which used telemedicine more frequently.16–18,20,23 One study reported that 77% of respondents would not have attended if the programme was not available via telemedicine.
20
Negative aspects of videoconferencing
Eight studies reported negative findings regarding videoconferencing.16,18–22,26,28
Technological aspects were reported as the main problem contributing to poorer outcomes in the studies. Comments on sound, picture and quality were often made. This included comments such as “patchy sound” or “poor camera placement”. Difficulty of use, unreliability and motivational aspects were reported in three of the studies. User training prior to implementation of videoconferencing sessions was recommended.21,22,26 Poorer interaction and rapport was also a factor in three of the studies.18,19,28 One study found that the majority of videoconferences were not useful as a resource to assist with patient care, with poor uptake amongst physicians.
26
In another study, concerns were expressed that telehealth would replace onsite staff and all education activities in remote locations.
22
Limitations
Limitations of study methodologies were reported in six studies. These included poor heterogeneity of groups, small sample sizes, poor follow up of data to indicate if practices changed in rural areas, poor response rates and biased opinions in the data.17,18,21,24,25,28 These limitations are important in relation to the present review.
Discussion
The present systematic review of the perceptions of physicians using videoconferencing methods for clinical supervision and educational support found that the technique was viewed favourably by physicians and medical students in a variety of settings. The main positive aspects associated with the use of videoconferencing were increased access to continuing medical education, clinical support and improvement in patient care. These findings were viewed more favourably by rural and remote physicians. Despite these general themes, only one study focussed on direct supervision for junior doctors. Most studies focussed on education and training purposes, with little mention of clinical supervision. Therefore, there is a clear paucity of literature in this field, indicating a need for further research.
The main negative findings common to all studies were technology problems associated with videoconferencing. These included poor quality equipment, particularly for remote areas. Poor uptake was also associated with difficulty in use, poor reliability and lack of training prior to implementation. It was suggested that these problems could be overcome through training and troubleshooting. This information is not new, as these problems have been reported in earlier studies. For example, bandwidth limitations in rural areas are known to restrict the use of videoconferencing. In some situations, Skype has been used with limited success, but more work is needed to overcome the related infrastructure problems.
The literature was divided as to whether the use of videoconferencing helped in rural staff retention. In one study, access to telemedicine was reported as an unimportant consideration for recruitment and retention when compared with other aspects such as relationships, workload and community factors. 17 Conversely, another study reported that participants felt that telemedicine would improve retention. 22 This difference of opinion suggests that further research is required, particularly if this information is used to inform government funding.
It is also evident that studies have not considered measures of cost effectiveness. While cost effectiveness has been measured in many patient-based studies,29,30 there were no clear indications that providing educational support or supervision via videoconferencing changed the costs to health services. This also highlights the need to measure cost-effectiveness in future clinical supervision studies.
A previous systematic review of methodological design in telemedicine found that many studies lack rigorous control, which can create ambiguity. However, it was also recommended that studies need naturalistic methodologies in order to capture important data. 31 In the present review, varying methodologies were employed, with information captured from both qualitative and quantitative techniques. This is essential, as important information may not be represented in quantitative studies alone. Perceptions, in particular, can only be captured by qualitative data collection. Perceptions are important for future research and therefore the incorporation of qualitative and descriptive results adds value to the review.
All of the evidence available was from studies which had relatively low levels of evidence according to the NHMRC guidelines. 15 This can introduce biases that can lead to over and underestimates of satisfaction and acceptance. 32 The large variability and inconsistency in the methodology raises questions about the rigour of the studies reviewed. Although most studies included Likert scales, these measured different aspects of physicians’ perceptions, and in varying contexts. Therefore, it was difficult to compare studies closely with one another. Many of the studies relied on small sample sizes which can also decrease the validity of the research. The differing demographics and poor heterogeneity of participants amongst studies also needs to be mentioned. In each study there were varying levels of experience from clinicians, differences in proximity to a major city centre and professional streams. This could lead to biased results and differences of opinion about the level of support required by physicians. This bias may be accentuated for clinicians in rural areas where there are often extremes in clinical experience.
Outcomes and findings were also measured and reported in a variety of ways, with emphasis on different perspectives among clinicians. The outcomes were not measured in a standard way across studies, which meant that a meta-analysis could not be conducted. However, this is not surprising considering the area of research and the need to capture perceptions through naturalistic methods.
A systematic review of videoconferencing for clinical supervision and educational support enabled us to identify the state of current knowledge in this emerging field. However, the strict inclusion and exclusion criteria meant that not all research pertaining to videoconferencing could be captured, which may represent a limitation of the review. Nonetheless, it clearly identified the gaps in knowledge and areas for further research.
Conclusion
Improving the retention of rural physicians has been a primary goal across the world. Provision of clinical supervision, training and education via videoconferencing has been achieved with successful outcomes for patient care, cost reduction and health worker stress. The present review found an overall positive perception and satisfaction amongst physicians using videoconferencing for these purposes. Videoconferencing was particularly useful in rural areas for both clinician education and patient care. Negative perceptions are commonly associated with technical and training aspects of the technology. There have been no studies of cost-effectiveness in using videoconferencing for supervision or educational support.
The present review highlights the paucity of literature on clinical supervision and educational support of physicians using videoconferencing. In particular, there are few studies in relation to clinical supervision in rural areas. Many studies included other health professionals or patient care models which did not report supervision needs. The review also found a wide variety of methodologies, which emphasises the need for a more rigorous study design. However, the use of qualitative and descriptive methods is also a strength in this research area, capturing important information. It is recommended that supervision of rural doctors via videoconferencing be investigated further, particularly for future guideline implementation and health policy development.
Footnotes
Acknowledgements
We thank the School of Medicine and Dentistry, James Cook University for support throughout the project.