• 20 Aug 2019
  • 15 min
  • 20 Aug 2019
  • 15 min

Jo Cheah interviews Melissa Baysari about the pros and cons of electronic prescribing systems in the hospital setting. Read the full article in Australian Prescriber.


Hi, I'm Jo Cheah and this is the Australian Prescriber Podcast. Today we'll be talking to Melissa Baysari who is an associate professor in digital health at both the University of Sydney and Macquarie University. Melissa, thanks for joining us.

You're welcome. Thanks for inviting me.

No worries. So just over the weekend you gave a keynote lecture at a digital health conference in Sydney. How did you go?

It went well, very well received. Lots of laughs and questions, which is good.

That's always a good sign.


So I'm sure a lot of our listeners would have had some exposure to computerised prescribing, whether it's from the planning and rolling out phase to the end user stage. So we'll jump right in. So what is computerised prescribing and how and why did it come about?

So in a hospital setting, I guess computerised prescribing replaces the inpatient medication chart. So typically the doctors will prescribe directly into the computerised system and pharmacists will review the orders from the system and then nurses will administer from the electronic system. So it's a whole kind of medication management process which used to be done on paper and is slowly being replaced by electronic prescribing systems. Your question about why it came about, I think in the US there have been a lot of benefits which have been shown to result from implementation of these systems. And so I think here there's been a real push to try and get these systems in place so that we can also achieve these benefits.

So while electronic prescribing systems have reduced some medication errors such as those from legibility or inaccuracies and things like that, they have definitely introduced some new errors. So what are some of the most common new errors that we're seeing?

Well definitely the most common is selection errors. So these are when people choose an item unintentionally from a drop down menu - so the item above or below the item that they would like to select. These are probably the most common, new type of error that have come about from implementation of electronic prescribing systems.

But there are all sorts of other errors. For example, editing errors. So when someone selects a medication order and then realises part of the order is incorrect and then goes to edit the order, they make an error in that process. Another common one is failing to change your default times. So often these systems, when you select a medication order it will default to a particular administration time and then it's up to the user to assess whether this time is appropriate and make the change if it's not. And people often forget to do that. So you could go in to say, okay, prescribe an antibiotic and it defaults to 8:00 AM. It's currently 3:00 PM and you'd like the dose to be given now. If you don't go in and change the time, the patient will receive the dose at 8:00 AM the next day. I guess people have to become very vigilant in using the system and double check things before they hit the final okay button.

Yeah. And apart from being vigilant, are there any other prevention mechanisms in place?

So from the user's perspective, probably not. I mean, I guess attend training, that's very important. And if possible, refresher training is good. I mean these systems are highly complex, so I think learning everything in one go is very challenging and that's normally what happens. So a user will attend an introductory session on what to do and then they're left on their own really to figure their way through the system. I think if possible to then attend additional training sessions to learn about the extra functionalities and parts of the system would be very useful for people. From an organisation perspective, I think there's a bit that can be done. So obviously training is important. Support on the wards if people encounter problems. And then there's a lot that can be done in design. So I think if the systems are designed well, then these new errors can be reduced.

So on the topic of training, do you have any tips for health professionals? You mentioned for doctors, nurses, pharmacists. Or what about any tips for the trainers? Often they're people who have shown interest in the area that want to get involved with digital health.

Okay. So maybe from a user perspective, I guess keeping in mind that every system is different and even the same system implemented in a different hospital is likely to be different. So because you use system A in hospital X and then you move to hospital Y and you still use system A, it might be slightly different. So I think as a user you should keep that in mind. And that goes for the trainer, as well. And I think there's this kind of tendency ...you know… have you heard the term digital natives? Like people who grow up with digital health. There's a tendency for people to think, okay so these people are digital natives. We don't really need to give them as much training as we would someone from an older generation who hasn't really experienced much IT. But I don't think that's true. These systems are very complex and everyone needs to get fully trained to be able to use them.

I haven't heard the term digital natives before. So in regards to ... you did mention before about the organisation having a say in regards to ensuring that there's adequate training and support on the wards. So what can organisations do when there are changes and updates to these systems and how can they effectively make sure that all that information is passed on to their users?

Yeah, that's really tricky. I guess it's how do organisations distribute information about a change in policy, for example. Or you know, and it's kind of the same thing really. It's how they communicate with their staff. And this is just another form of information that needs to be distributed. So if a change is made to a system or an update is made to a system, I mean, I think a typical approach is probably that everyone gets sent an email with a summary of how their system has changed or what they need to do differently. And whether people log into their email and read that information, I don't know. I mean we probably need to think of more creative ways of communicating things to users. I mean information sessions are very common, like through for example, grand rounds or regular education on the wards and so on. So I think different user types also probably respond differently to different educational methods. So we did a study recently where we looked at vancomycin and we looked at the education that prescribers received on vancomycin and nurses received on vancomycin and pharmacists. And we found that there were differences depending on the type of healthcare provider on what they responded to best. So for example, nurses were really procedural and liked to get steps provided to them on what they needed to do. The junior doctors were really happy with face-to-face training, but the senior doctors preferred to just be sent the information out, that kind of thing. So I think we'll probably need to think about what works for different people and provide a range of different options so that people can respond best to what works for them.

We were talking about the inpatient medical chart before. So prior to using electronic prescribing systems, we did have a standard national inpatient medication chart.


So are there any national standards for the design of electronic prescribing systems? For example, in the way they look or their symbols or functions?

There are actually national guidelines that we have here in Australia for electronic medication management systems. So they're put out by the Australian commission on safety and quality in healthcare. So they have national guidelines for the onscreen display of medicine information. And these include I guess guidelines on what you should do for displaying medicine information in an electronic prescribing system. So things like for example, you should use full English words, don't use symbols. You should leave a blank space between a number and the unit. So these kinds of things. So when people are designing systems, they should really abide by these guidelines to ensure that the information is read and processed correctly. So the Commission has quite a few guides on its website for the implementation of these kinds of systems. Because the system design itself is one thing, but the way you implement the system is another. So they have a guide for the safe implementation of electronic prescribing. And so things like that you need the support, you need the training you need champions to drive it, et cetera.

And how do you feel about different networks or different hospitals rolling out different systems that may not talk to one another?

I mean I think in an ideal world everyone would use the same system and every system would talk to one another. But I think it's very challenging. There are a lot of, I guess, factors that go into deciding on which system to take. I'm a little bit biased because my area of expertise is human factors and I wish that usability played a larger role in the selection of systems, but I think that's kind of left to the sideline. And I think it's much more about functionalities of what the systems propose to do, which is important. So yeah, I think people tend to work in silos like anywhere really. And I think the conference, for example, on the weekend was a good opportunity for people to come together that have implemented different systems so that we can all kind of learn from each other. And I think those kinds of forums are very important for people in different areas to hear about different systems, what works, what doesn't. And it's amazing even with the different systems in place, the challenges that people experience are very similar.

So once a system is rolled out in a hospital or health centre, we obviously have to review and audit how well it's going. So for people who are doing that sort of job, the reviewing and auditing, what sorts of things should they be looking out for?

I guess it depends how many resources you have. How much time and effort you can put towards evaluation. I think evaluation is very important and I think it's probably not given enough attention. I think new types of errors is a big issue and people should focus on that. When a system first goes in, there's, I think, there should be a lot of monitoring in terms of like glitches in the system. I mean testing is done before system goes in, but often things happen when it's implemented on a larger scale. And gaps in training, those kinds of things should be identified right away. So I think for even for the first six months, things should be monitored very closely because there are a lot of unexpected consequences which emerge following implementation.

So talking sort of larger scale now, as you mentioned, in America this is sort of more advanced than how we're going in Australia. So can you tell us a little bit about the research that's gone into looking at the effectiveness of these systems in regards to medication safety.

Okay, yep. So there have been quite a few trials now that have shown that implementation of electronic prescribing reduces medication errors. The most common type that has been explored are prescribing errors. Whether they impact on other error types, the evidence is kind of mixed. So whether they impact on, for example, medication administration errors that nurses make. I know there are some studies have shown that they do and some studies have shown that they don't. Also most of the research kind of stops on the medication error thing and doesn't go further to look at the impact that these systems have on patient outcomes, like harm associated with those medication errors. So for example, the system results in fewer wrong doses. Does this then translate to less patients being harmed by wrong doses? We don't really know. In Australia, we have done some trials that have shown that electronic prescribing systems reduce medication errors, specifically prescribing errors and at the moment we're running a large trial to go further than that to look at patient harm. With the implementation of these systems, everyone expects there to be benefits in safety and accompanying that often there's increased work. So, for example, on a national inpatient medication chart, a doctor can complete these very quickly and it's not really ... although all fields are meant to be compulsory, a field left blank here and there is not really going to be picked up or changed. In an electronic system you can force the prescriber to fill in every component of an order. And so this therefore takes more time. And so often users think about the impact that the implementation of these systems have on how much work they need to do. So there have been some trials here in Australia where time and motion studies were done to look at the impact of these systems on work. And what they've shown is that that's actually not the case. So doctors do not spend less time with patients. There often is a redistribution of work, so they might have to spend more time writing a prescription, but then they'll spend less time doing some other medication task because the prescription is now nice and clear and they won't need to follow up so much with other people, for example.

Sounds good. So Melissa, that's all the time we have for today. Thank you so much for joining us on the Australian Prescriber Podcast, and thank you to my good friend and a couple of my colleagues who let me pick their brains about computerised prescribing before this interview.

The views of the hosts and the guests on the podcasts are their own and may not represent Australian Prescriber or NPS MedicineWise. I'm Jo Cheah, thanks for listening and I'll catch you next time.