Systematic review
Skills laboratories: Virtual training of manual and practical skills
Systematic review
|Updated
The National knowledge centre for the health services was given an assignment by the Spesialitetsrådet, The Norwegian Medical Association, to summarise available research on the effect of virtual training of skills by using models or different categories of simulators which are used in specialist training of physicians.
Key message
Virtual training of manual and practical skills in skills laboratories gives possibilities for training without the involvement of real patients by use of physical models, or different types of simulators. Virtual training can be used for development of component skills or as a full scale simulation to practice an entire procedure in a realistic environment, but without living patients. The National knowledge centre for the health services was given an assignment by the Spesialitetsrådet, The Norwegian Medical Association, to summarise available research on the effect of virtual training of skills by using models or different categories of simulators which are used in specialist training of physicians. The report is based on the results of six systematic reviews that were published within January 2008. All of the reviews deal with training of surgical skills. We have not identified any systematic reviews from other medical fields where virtual training has been used, despite growing number of primary studies. Summed up research shows:
- Virtual training of skills without the use of real patients, but with models and simulators, given in addition to standard training, may improve surgical skills and may increase general competence.
- Surgical skills practiced in skills laboratories or simulator training can probably be transferred to clinical practice, but documentation is still limited and the results should be interpreted with care.
- The research does not give grounds for conclusion about which type of virtual training is most effective or how training should best be organised.
- The quality of available research is too low to decide if virtual training of skills has an impact on patient oriented outcomes such as pain, mortality, morbidity or satisfaction.
- The impact of virtual training of skills with simulators and models is best documented with regard to training of certain technical and manual skills.
- The impact of virtual training is still poorly documented with regards to whether this training actually improves the quality of health services and improves patient security.
Summary
Background
The introduction of highly technological equipment in medicine brings with it a need to train new skills, and following that, the need for training to maintain a necessary competence level. Virtual training in medical education opens possibilities for training of manual technical skills and nontechnical skills like teamwork without the use of real patients. Virtual skill training can be divided into several different categories:
• Physical synthetic models with direct visual evaluation
• Box-simulators with or without video evaluation make possible training of skills on organs using artificial organs made of different synthetic materials that imitate human tissue
• Computer based simulators composed of a computer with software and a control unit to move instruments and perform procedures (microsimulation)
• Computer based simulation with electronic mannequins with highly sophisticated electronic equipment, often connected to an operating procedure and video technique (macrosimulation)
•Animals (live anesthetised animals or animal cadavers) or organs and tissue from dead human beings
National knowledge centre for the health services was given an assignment by the Spesialitetsrådet, The Norwegian Medical Association, to summarise available research on the effect of virtual training of skills by using models or different categories of simulators which are used in specialist training of medical candidates.
Methods
We searched for studies in Medline, EMBASE and Cochrane library (CDSR, DARE, HTA, NHSEED) published before September 2008. The search was limited to systematic reviews of studies with mixed study designs, but with a larger portion of randomised controlled studies (>50%). The participants were medical doctors and medical students. We included interventions such as training in skills laboratories and different categories of simulators including training on animals. We have registered the following outcomes: improvement in skills, patient oriented outcomes, procedure oriented outcomes, transferability to the clinic and costs/costefficiency. Two researchers evaluated the quality of the systematic reviews and retrieved the results independent of each other. We used the GRADE system to evaluate the quality of the evidence for the specified outcomes (10;11).
Results
We identified 1386 unique titles in our search for systematic reviews and 58 potentially relevant references were ordered in full text. Eight of these fulfilled the inclusion criteria, of which two were protocols for ongoing Cochrane reviews, and one was an update of an earlier report. In Desember 2008 Gurusamy et al published their results and the Cochrane review was published in January 2009. We decided to add this review to the report because of its relevance. The report, therefore, builds on six systematic reviews all together. The effect of the interventions in all the studies was measured against the control group which received standard training or no training.
We have grouped the results according to intervention and type of outcome.
• Video box training vs. no training or standard training in laparoscopic surgery
The studies on effect of video box training on surgical skills measured as the ability to execute predefined tasks or general accomplishment level gave inconsistent results. Over half of the studies showed no significant difference between the video box training group and the control group, and only a third of the studies showed significantly better effect than no training or standard training. There was no significant difference between the groups with regards to the number of reported procedure errors or procedure time. We do not know if the trained skills can be transferred to clinical practice.
• Computer based simulator training vs. no training or standard training in laparoscopy, endoscopy, colonoscopy or sigmoidoscopy
The majority of studies showed a significant change in skill measured as an ability to perform predefined tasks, general performance level, and as a global scoring rate after training with a computer based simulator compared to the control group with no training. Compared to surgeons with the standard training the difference in effect was small or not significant. Evaluations done on simulators, live animals (pigs), or even on real patients have shown that this competence was transferable to the clinic. Reported results for procedure related outcomes and patient oriented outcomes were ambiguous, and do not provide grounds for a conclusion.
• Computer based simulator training vs. video box training of general surgical skills
The results were ambiguous. Half the studies showed a better training effect in the intervention group, while the other half showed no significant difference. One study showed better effect in the control group. There was insufficient data for an evaluation of differences in effect of training of procedure related out comes such as time, errors and injuries. None of the studies reported of transfer ability of the skills to the clinic. Identified documentation does not give grounds to conclude if computer based simulator training is better than video box training.
• Training with surgical models vs. no training or standard training of surgical techniques
The results in the intervention group were significantly better in four out of six studies. Two studies showed transferability of the trained skills to the clinic in execution of tasks on pigs. The same studies showed a shorter procedure time in the intervention group compared to the control group.
• Computer based simulator training of a team in trauma treatment and Advanced Cardiovascular Life Support (ACLS)
Results showed better general performance of tasks, increased knowledge level and security in the intervention group compared to the group who had been given standard training in two of the three studies in total.
Discussion
Good patient security is a goal for the health services and a demand from society, and this strengthens the need for safe training of skills in medical doctors in specialist clinical services. This report is based on six systematic reviews of which five of them focused on the effect of virtual training of surgical skills. This means that the generated knowledge does not cover all the areas of medicine where virtual training of skills is used. We have found no systematic reviews of studies on the effects of training of nontechnical skills or team training. Examples of such fields are gynaecology, anaesthetics, trauma treatment or Advanced Cardiovascular Life Support.
The limits to systematic reviews also means that results from newer studies are not included. Most of the studies reported an effect of skill training using models or simulators developed for training for surgical skills in peephole surgery. There is a lack of summarised research on effects of virtual training in other medical fields, but the number of primary studies is growing rapidly.
There is also a need for an objective evaluation of skills. Changes in skills were in most studies evaluated using a simulator and not on patients. Such an objective evaluation of skills gives a good indication on skill change on a simulator, but does not show if these skills are transferable to the clinic. The quality of available research was too low to decide if virtual training influences patientrelated outcomes such as pain and discomfort, mortality and morbidity or satisfaction. The research does not give grounds for conclusion about which type of virtual training is the most effective. Future research may alter some of the conclusions. A summary of the existing research shows that most of the important questions that are relevant for clinical practice are not studied.
Conclusion
• Virtual training of different surgical skills on models or simulators in addition to standard training can improve surgical skills and increase the general competence of the surgeon.
• Surgical skills trained in skills laboratories or simulator training can probably be transferred to clinical practice, but documentation is still limited and should be interpreted with care.
• The impact of virtual skill training on the quality of health services and patient security/safety is still poorly documented.
• We have found no summed up research on effects of virtual training in other fields besides surgery.