Sunday, December 18, 2011

Distance Education can be Effective for Rural Physicians

Rural doctors don't have the same access to learning resources as physicians in cities often do, they depend much more on distance learning methods than other physicians do. We know that distance learning methods can be just as effective as any other methods. Hepatitis C is a problem in rural New Mexico. The treatment is complex, and rural physicians often don't have access to effective continuing education to help them learn to treat it. Physicians at the University of New Mexico have developed an interesting system for using a telehealth network to teach Hepatitis C treatment, called Project ECHO.



In the Project ECHO model, physicians begin by spending a day or two at a Hepatitis C clinic, where they shadow a specialist.  Then, follow weekly, two-hour case presentations that include discussion of treatment complications and psychiatric, medical, and substance abuse issues.  Collaborators in these sessions might include various medical and mental health specialists. 

The cases are worked through collaboratively, with shared case management and decision making.  The rural physicians present their own cases, and occasionally, are asked to research and present didactic information to the group. 

Project ECHO has resulted in positive differences in the lives of patients.  The outcomes for the ECHO project, published in a June, 2011 article in the New England Journal of Medicine, showed that patients treated at the ECHO sites had similar rates of sustained viral response as patients seen at the specialty clinic.

Sunday, November 27, 2011

CME Associated with Decreased Mortality in Patients with Coronary Heart Disease

Low density lipoprotein
Evidence showing real changes in patient outcomes are rare, even though that's the ultimate goal for most CME programs.  A very nice study, published in the Annals of Family Medicine, showed a change in practice, and a positive change in patient outcomes after a CME program (Kiessling, A.; Lewitt, M.; and Henriksson, P. (2011), Case-based training of evidence-based clinical practice in primary care and decreased mortality in patients with coronary heart diseases.  Annals of Family Medicine, 9:211-218.).   

CME on new Guidelines for Coronary Heart Disease
Practice guidelines in Sweden on the use of lipid lowering drugs were developed after the 1994 publication of the effectiveness of lipid lowering in coronary heart disease.  In order to help general practitioners follow the new guidelines, a series of CME programs were developed. The training was designed to be interactive and to "activate" learners.  They felt that a case-based method was the most effective way to activate the learners and help them improve the sills and attitudes needed for decision making. 

CME Design
All participants were mailed a copy of the new guidelines.  The instruction began with a lecture deliverd by a specialist.  This was followed by a series of seminars, delivered over the next two years.  Physicians participated in 1-2 cases each year, in groups of 4-7 people.  The cases were designed to be simple and well-organized, and presented a problem without an obvious correct answer.  The seminars were one hour each and consisted of the case presentation, followed by interactive problem-solving, with a thorough discussion of the pros and cons of various decisions.  Physicians in the control group only listened to the lecture. 

Results
This seemingly simple design resulted in changes in practice, that at two years, had significantly reduced patient's low-density lipoprotein cholesterol levels compared to a control group. 

Ten years later, the mortality rate of the control group was 44%, while mortality in the intervention group was 22%.  That's right, ten years later, patients whose doctors went through a few interactive cases with an specialist,  had half the mortality of patients whose doctors listened only to a lecture and read the guidelines. 

What Happened?
The results of this study are pretty amazing.  Long -term effects are very hard to find in education studies.  It's very rare to see a study looking at an effect over a few months, let alone years, but there are a few things that we see with other kinds of successful CME that may be giving us a hint about why this worked so well. 

  1. Learning over a period of time.  It takes time to learn new skills and new habits.  Physicians in this study had two years of exposure to this material.  This repeated exposure may well have helped these doctors incorporate the new guidelines into their daily practice. 

Interactive and problem focused.  The cases were designed so that they didn't have clear or obvious answers, and participants were given the chance work their way through all of the nuances of these cases, including "defining and valuing the pros and cons, and feasibility of different decisions."  One of the problems with presentations by experts is that they don't let learners work through these important details themselves. 

Sunday, September 18, 2011

New Review of Simulation Education

A new article in JAMA looks at the effectiveness of simulations in health education.  The meta-analysis, which looked at over 600 published research papers, found large effects for the use of technology-enhanced simulations for knowledge, skills, and behaviors. 

Many of the papers included in the analysis dealt with surgical procedures, but there were other studies too, including dentistry, communication skills, nursing skills, examination skills, and obstetrics. 

The big limitation with this analysis is that it only looked at studies comparing simulation to no-instruction controls.  It's not surprising that some kind of instructional intervention is better than nothing at all, but this paper can serve as a good resource for those interested in learning more about simulations in medical education. 

Sunday, September 11, 2011

Interactive CME Improves Physician Communication Skills

A very nice study, published last year in Patient Education and Counseling, shows that interactive instruction, that's heavily based on practicing skills, can have a positive influence on physician practice.  The study uses several elements that have been shown to influence provider behavior: skills practice, interactivity, and a design that spreads over several different sections (spaced learning). 

The Practice Setting
Physicians in a university health service clinic where there is a strong need to counsel patients on risky behaviors. 

The Instructional Design
The instruction consisted of five 2-hour workshops at 4-6 week intervals, over a span of six months.  the workshops all emphasized communication skills, and each focused on a specific area of need including managing difficult patient encounters, screening and assessment for depression, screening and assessment for alcohol use, taking a sexual history, and behavior change counseling.

Each of the workshops began with a short video or demonstration, to trigger discussion, then there was a brief, highly interactive lecture.  After that, the majority of the time was spent practicing skills with standardized patients and receiving feedback.  After each session, participants were given pocket cards to uses as references, and readings for further information. 

The Measures
This very thorough study, measured four different outcomes - participant satisfaction, learning (measured prior to the intervention and one month afterward), performance (assessed by chart review both before and after the intervention), and the impact on patient satisfaction.

The Results
The results showed positive increases in many important outcomes including knowledge, and more importantly, physicians changed their practice.  The patient satisfaction scores were already very high, and did not increase significantly as a result of the training. 


What it Means
Providers can change their practice.  Often, change requires something more than just education and training, but the elements that can help physicians change practice behavior are not a mystery, we see the same elements over and over again when we find CME that that changes practice:

  1. Skills practice with feedback - people need to practice new skills before they can use them
  2. Interactive information delivery
  3. Practice over time - time is an often overlooked factor, but one of the most powerful

Sunday, August 14, 2011

Effective Biology Teaching

We know that people don't necessarily learn much from passive methods like listening to lectures or reading.  People need to actively process the knowledge, practice using it, and get feedback. 

It's not always clear how to do that though.  What can people do to increase the kinds of activities that lead to better learning outcomes?  We can look to some of the work being done that looks at learning in other disciplines.  I've been very excited by some of the work that's being done in science, technology, engineering, and mathematics (STEM) teaching.  There have been a lot of recent developments in STEM teaching, based on what we know about the science of learning. 

I recently came across an excellent review from the Annual Review of Cell and Developmental Biology (2009, W.B.Wood, Innovations in teaching undergraduate biology and why we need them.  Ann, Rev. Cell Dev. Biol., 25:93-112).  This article very nicely lays out the argument for better teaching methods, and gives some very practical advice on how to implement them. 

Some Key points are:

Recent research in educational psychology, cognitive science, and neurobiology has yielded important new insights into how people learn and the optimal conditions for learning.

Discipline-based educational research (DBER) has led to the development of teaching approaches based on these insights (promising practices) and has provided extensive evidence that these approaches can be substantially more effective than traditional lecturing even in large classes.

These promising practices vary in their ease of implementation but even their partial adoption can lead to significant gains in student learning.


The recommendations would work well for teaching in many different settings, including continuing medical education.  Teaching recommendations include:

Effective instruction must build on students’ prior knowledge (which may include misconceptions that require correction).

Instructors should be aware of the student diversity in their classrooms and use a variety of teaching modes to optimize learning for all students.

Classes should include frequent formative assessment to provide feedback to both instructors and students.

Students should be encouraged to examine and monitor their own understanding of new concepts, for example, by explaining these concepts to their peers.

Students should be encouraged to work cooperatively and collaboratively in small groups.

In order to bring about the neurological changes that constitute learning, students should spend time actively engaged with the subject matter, for example, discussing, diagramming, solving problems, working on a research project, etc., in addition to or in place of listening passively to a lecture, reading the textbook, or consulting Web sites.

This paper is highly recommended for anyone trying to learn more about how learning science is being applied today. 

Sunday, August 7, 2011

ACCME Interviews


ACCME has been producing series of video interviews about continuing medical education.  The interviewees are experienced practitioners who share some of the things they've learned while producing CMEs. 

In June, they interviewed Rick Kennison of Peer-Point Medical Education Institute.  He's doing a lot of work integrating the CME process into an improvement framework that impacts patient outcomes.  Sometimes people think of education as the "solution," but a lot of times, education works best if it's part of an overall improvement effort. 

Sunday, July 17, 2011

Internet-Based Learning Doesn't Take less Time than Classroom Learning

Does internet-based learning take less time than classroom learning, as claimed by many authors?   Probably not.

Cook, Levison and Garside, (2010, Advances in Health Sciences Education, 15(5):755-70), looked at studies that compared the amount of time spent learning a topic in the classroom and over the internet.  They found that internet-based methods not necessarily any more efficient than classroom-based  In fact, if you use methods that increase learning, like interactivity, internet-based learning may take longer. 

"... our data suggest that there are few shortcuts to learning.  In this review, nearly all of the modifications of IBL [internet-based learning] to improve learning outcomes (interactivity, feedback, audio, video, etc.) required more time on task."  

Internet-based methods can be great time savers if you're at the cost of getting to a classroom, but learning takes time,  no matter how it is delivered.  

Sunday, June 19, 2011

Active Learning Beats Lectures in a College Physics Class

CME presented as didactic lectures don't change practice behavior, and most of the time, they do little to improve knowledge.  So why are they used so often?  I think people use lectures partly because that's the way they learned in college, so they think it must work.  The problem though, is that college is a completely different situation because college students have to recall information on tests.  This means that they often use a variety of techniques to help them remember the information in a lecture.  Professionals don't have to pas high stakes tests (at least not for most CME), so they won't go through that extra work. 

Lectures aren't always the best way to learn, even in college.  A recent article in Science (hat tip to Donald Clark) compared learning in a physics class sections taught in two different ways.  The first was taught by an experienced, highly-rated lecturer, and the second, was used methods developed from learning theory, and was taught by a graduate assistant with no teaching experience.   This class used a variety of interactive methods based on the idea of deliberate practice.
"Deliberate practice takes the form of a series of challenging questions and tasks, that require the students to practice physicist-like reasoning and problem-solving during class time while provided with frequent feedback." 
The goal was to have students spend their time learning to reason and think like scientists, rather than simply being given facts.  It's important to note that both classes used clickers, and even used many of the same clicker questions, but the lecture-based class used them more like a post instruction assessment, rather than a way to start discussion and keep it going. 

The graduate assistant didn't lecture, he presented the problems, and gave guidance and explanations.  He also responded to student's answers for the clicker questions and to things that he heard during the student discussions. 

The results were uniformly positive.  The active learning group had higher attendance, higher levels of engagement, and most importantly, twice the learning of the group that received the lectures. 



Active learning, centered around problems can easily be incorporated into CME activities.  In fact, given the fact that these active learning experiments are more enjoyable for students, and more likely to result in better learning outcomes, it's hard to come up with reasons not to at least try them. 

Louis Deslauriers, Ellen Schelew, and Carl Wiemann (2011), Improved Learning in a Large-Enrollment Physics Class, Science 332:862-864.  http://www.sciencemag.org/content/332/6031/862.abstract

Sunday, May 22, 2011

Review of Effective Internet Interventions for Health Education

We know that internet-based education can be effective for health education, a recent review of internet-based health care instruction (2010, Academic Medicine 85:909-922), suggests which kinds of materials may be the most effective. 

The authors searched several databases for papers looking at the effect of internet-based learning for a number of different audiences including students and professional doctors, nurses, pharmacists, dentists, and others.  They classified the different approaches to get an idea of what kinds of different designs lead to the most effective learning outcomes.

They found that the methods that lead to better learning outcomes include interactivity, practice exercises, repetition, and feedback.  Interactivity included activities such as self-assessment questions, interactive models, or thought-stimulating activities.  Repetitive designs used repeated exposures to materials, for instance, emails with multiple repetitions of the same material. 

This paper is freely available and is a very good read for people interested in a survey of the different online methods used for health education, and for a review of the methods that have proven the most effective.

Sunday, May 1, 2011

Is Classroom Learning Better?

Distance learning is often cheaper and more convenient than classroom-learning.  Sometimes people have a feeling though, that it just can't be quite as good as classroom learning -  We must be giving up something if we try to teach people at a distance. 

Distance and internet learning tends to hold up in studies comparing it to classroom based learning.  A recent paper in JAMA looked at increases in knowledge and practice change for education on the NIH Cholesterol Education Program, comparing distance learning and live-CME, and found little difference between the 2 methods (with the only difference favoring learning at a distance).  (JAMA, Sept. 7, 294:1043-1051). 

This is a nice study, because it compared distance methods to live-CME methods that had been shown to be successful in the past.  The two instructional methods were designed to be as similar as possible.  Both had a one-hour didactic session, followed by interactive cases with feedback, supporting resources (like a risk calculator), step-by-step clinical practice guidelines, a guidelines summary, and access to expert advice.  Students in the live-sessions attended a 1.5-2 hour class.  Student using the online instruction had 10 days to complete their materials. 

The researchers took a thorough look at instructional outcomes by measuring knowledge before the interventions, immediately after, and most important, 12 months later.  Practice change was assessed by chart audits. 

There was only one significant difference between the two groups, the internet group had a significantly higher number of high-risk patients treated with pharmacotherapuetics according to the guidelines.  Internet-based CME was effective. 

This study isn't out there by itself, there have been many studies comparing distance methods to live instruction.  It's very common to find that well-designed distance methods are usually as good, and sometime slightly better, than classroom instruction, a finding that is reinforced by a recent report by the US Dept. of Education goes into great depth comparing outcomes in distance education and comes to a similar conclusion. 

This is an area that needs more study because it's likely that classroom-based learning will be more efficient for certain learning objectives and learners, but for many skills, distance learning has proven just as effective as classroom training. 

Sunday, April 24, 2011

SImulations Can Work Well in Medical Education.

There are lots of ways to teach things, but we know that to develop real expertise in a skill, they need the chance to practice and to get feedback.  One way that people can get that practice and feedback, is by simulating the work environment.  Simulations are being used more often in medical education, and they've proven to be effective, if done correctly. 

A recent review of simulations in medical education looks at some of the aspects that make learning from simulations effective.  They include:

  • Feedback
  • Deliberate practice
  • Curriculum integration
  • Outcome measurement
  • Fidelity to the work environment
  • A mastery learning environment
  • Transfer to practice

These are the usual suspects in effective learning environments - especially practice and feedback.  We should try to incorporate these elements as often as we can when we're designing learning environments for skill learning. 

A Critical Review of Simulation-Based Medical Education Research: 2003 - 2009, Medical Education, 2010, 44:50-63. 

Wednesday, April 6, 2011

Teaching and Learning in the Year 1011

Learning 1011 AD and 2011 AD, courtesy of Nathan Wallen
Let's go 1000 years back in time to set up a learning event. How would you do it? You would probably bring the expert, and the students into the same room, have the students sit on hard wooden benches, and arrange to have the expert stand in front and transfer his knowledge to the students. This kind of direct teaching was was one of the few options available, and it was reasonably efficient, which is why it was used for thousands of years.

Now, of course, we don't need to be in the same room with an expert to learn. We can interact with many different experts, with other learners, and with many different representations of the content. We can design learning environments and online spaces that help people practice, work through problems, and get feedback. If there's a need to listen to an expert lecture, we can record the lecture and let people view it whenever they want. And the good new is, we have a lot of data that shows that of these kinds of activities will result in better learning, and will be more likely to actually improve professional practice.

So, other than the hard wooden chairs, why does so much professional continuing education in 2011 look exactly the same as it would have in 1011?  It doesn't have to be that way, we can help professionals continue to learn and improve.  

Sunday, March 20, 2011

Pseudoteaching

The Action-Reaction blog has an interesting post on "pseudoteaching" at MIT. They have defined pseudoteaching as teaching that looks like good teaching, but pseudoteaching doesn't help students learn. Pseudoteaching that's entertaining may look like good teaching to everyone involved - the teacher, to the students, and to observers, even though it's not very effective.

The post highlights John Lewin, a highly regarded physics professor who "clearly loves physics, and he loves sharing it with his students." His lectures were carefully rehearsed, practiced, and extremely entertaining.

The problem? The failure rate in his classes were too high, and by the end of the term, only 40% of students were attending his classes.

"...Lewin was pseudoteaching. It looks like good teaching, but he was the one doing all the talking. It looks like the students are learning, but they were just sitting there watching. It’s like trying to learn to play piano or play a sport by watching your teacher or coach. It doesn’t work well."


MIT is now using a hands-on approach, called TEAL (Technology Enhanced Active Learning), which has students and teachers doing experiments together, and working through problems.

Unfortunately, much of the teaching that happens in CME is pseudoteaching: Lectures that talk at people, rather than helping them work their way through problems and cases. The real value that experts can bring to a classroom is not the facts they know, but the ways that they think through problems. We know that lectures don't work in CME, It's been shown time and again that lecturing to professional does not change practice behavior. It's time to use more effective methods.



Lewin's lectures look very entertaining, but entertaining students, and actually helping them learn are not always the same thing. This is what MIT has moved away from:

Sunday, March 13, 2011

More Positive Results for Spaced Education

In my last entry, I talked about the learning benefits of spaced education for 3rd year medical students doing a urology rotation. Spaced education means simply extended the learning and practice over time. The study I referenced looked at learning after giving students a series of questions using email and found better learning results for the students who had received this kind of education.

The authors of that study did a follow-up study (Journal of Urology, 2009, 181:2671-2673) and found that the benefits of that treatment could still be found, more than 2 years later. This is an amazing result for an educational study. It will certainly need to be replicated but it is very encouraging and speaks to the strength of spacing education out over time.

The authors found that students who had received the spaced education did significantly better on a test of the material than students who had not received the spaced education intervention. The test questions they were given were different from the questions they had answered in the original study (but were from the same batch so they were somewhat similar).

Spaced education and practice can be used fairly easily by using email or even mobile devices, a matter of spreading the practice over time. It should be used more often in CME applications.

Sunday, February 27, 2011

Learning from email

One of the things that I want to explore on this blog are the techniques that have been particularly successful in medical education and CME. In 2007, a study was published in Medical Education, that I find particularly interesting (Medical Education, 2007, 41(1):23-31.)

The paper looked at the effect of "spaced education" which has been a consistently strong way to deliver education and training. Spaced education is simply education that's delivered over time. Instead of an hour of straight learning, you might have 10 minutes of learning once a day for six days. Spaced education is powerful, results of studies over may decades have shown that spaced education is more efficient, and it's more effective than "massed" learning. It's perfect for the tools we have available today, such as email and mobile technology.

The authors of the study used a pretty simple intervention. Third year medical students, in a one week urology rotation, were sent 10 - 13 emails with multiple choice questions or simple cases related to their urology learning objectives. . They got the emails over a period of up to 11 months. At the end of the academic year, they took a test on the content.

The students who had received the spaced education did significantly better on the test, and the strongest effects were up to 11 months after the intervention.

I like this study and its simple design - it used a powerful, and tested, learning technique and paired it with a ubiquitous tool (email), to get a strong outcome. I think we're likely to see more spaced educational interventions in the future.

Monday, February 21, 2011

Guidelines for CME

In developing CMEs, I've always tried to make sure that the program incorporates, or at least mentions, any relevant clinical practice guidelines. There are, afterall, the best collective wisdom that we've got. That's why I was excited to stumble onto the Evidence-Based Guidelines for CME recently published in a special supplement of Chest (Chest, 2009, 135, Supplement).

There's lots of information for developers of CME programs, and for those researching CME effectiveness. The Guidelines cover the effect of CME on knowledge, practice performance, clinical outcomes, and include a discussion on audience characteristics and external factors in CME, and the use and effectiveness of simulations.

Sunday, January 30, 2011

What Works to Change Practice

What can we do to help physicians improve their practice? There is research that can help guide decisions and priorities. In a 1998 “review of reviews,” looks at methods that have improved practice. The educational message is pretty clear – didactic methods will not work, we need to include education that that helps physicians think about, and practice their new knowledge.




Consistently Effective Interventions

Reminders (manual or computerized)

Multifaceted interventions

Interactive educational meetings that include at least include discussion or practice



Interventions of Variable Effectiveness

Audit and feedback

Local opinion leaders

Local consensus processes

Patient mediated interventions



Interventions with little or no effect

Educational materials, including clinical practice guidelines, recommendations for clinical care, audiovisual materials, and electronic publications

Didactic educational meetings





Bero, L.A., Grilli,R., Grimshaw,J.R., Harvey, E., Oxman, A.D., and M Thomson (1998) Closing the gap between research and practice: an overview of systematic reviews of interventions to promote the implementation of research findings. BMJ; 317 : 465.

Monday, January 17, 2011

CME in Second Life

There was a big surge in interest in virtual reality as a training tool about five years ago when Second Life, and several other virtual reality tools became widely available.  Much of that initial excitement has died down.  I suspect that there are a number of reasons that it has faded but one reason may be that many people did not use the technology very well.  I went to a number of presentations in Second Life, but most were merely lectures, which can get boring, no matter what the technology. 

Lectures aren't the only things that you can do in Second Life though.  I think that one very promising use of the technology might be for training that involves role playing. Virtual reality offers some potential advantages over face-to-face role playing - including alleviating a little of the embarrassment that people can feel when they're the center-of-attention. 

Two recent blog posts (first, second) by Robin Heyden describe the experience of delivering continuing medical education on motivational interviewing to family physicians.  The training included a 40 minute presentation on motivational interviewing, followed by practice sessions. 

There are lots of caveats, including the usual problems that new user often have accessing all of the media in Second Life, but overall, it's a very interesting step and one that I'll be paying attention on the future. 

Here is the presentation portion of the training:

Sunday, January 9, 2011

CME can alleviate misconceptions about back pain

A recent paper in Spine (2009, 34(11), 1218-1226) looked at general practitioner knowledge about back pain treatments. 
Doctors who had recently participated in CMEs about lower back pain had significantly better pain management beliefs, supporting the importance of CME for updating people’s knowledge. 

One surprising result was that physicians who self-identified as having a special interest in lower back pain were actually less likely to understand the most appropriate treatments, which include continued activity, rather than bed-rest.   That’s right, physicians in this survey, with an interest in lower back pain, actually knew less about how to treat it.  This may be a statistical hiccup, but it might also fit with the idea that people are actually very poor at self-evaluating their own knowledge and skills (see, for instance Advances in Health Sciences Education, 2004, 9(3):211-24).