High-quality health records to support patient care

Brit Harper (BH): Welcome everybody. You're joining us for another session from the RACGP’s Practice Essentials Webinar Series. This webinar is ‘High-quality health records to support patient care’. My name is Brit and I'm a Project Coordinator with the RACGP’s Practice Technology and Management team. I'll be with you for the presentation. 

I'm joined by Dr Steven Kaye, a GP based in Melbourne, who will deliver the presentation today. Steven has been an RACGP Expert Committee member in eHealth and Practice Technology and Management since 2016. He's previously been the deputy chair of the Bayside Medicare Local and Chair of the Bayside GP Network. Steven is a GP registrar supervisor for Eastern Victoria GP training, an examiner for the RACGP Fellowship, and the managing partner of his general practice in South East Melbourne. He has worked on many projects in an advisory role, including the ‘Shaping a Healthy Australia’ working group and My Health Record education program, and the My Health Record Assurance Committee. Steven, welcome to the webinar. 

Dr Steven Kaye (SK): Thanks Brit and welcome everybody who is listening.

BH: Thanks, Steven, and thank you all for taking time out of your busy schedules to attend the session.

Before we begin our webinar, I would like to acknowledge the traditional owners of the respective lands on which we are meeting today and pay my respects to elders past, present and emerging. I would also like to acknowledge any Aboriginal and Torres Strait Islander people attending the session.

This activity is delivered by the Royal Australian College of General Practitioners or RACGP in collaboration with Commonwealth Scientific and Industrial Research Organisation or CSIRO. I'll note at this point that the education activity is accredited for two points under the RACGP’s 2020-2022 CPD triennium. To collect these, you do need to watch the whole webinar. The RACGP is working with the CSIRO on its Primary Care Data Quality Foundations Programme. There are several projects under this banner. One of these involves the development of a primary care data dictionary. We'll talk more about this at the end of the presentation.

I'm going to take a minute now to discuss the features of GoToWebinar. If you’ve joined us for webinars in the past, you'll notice that your screen looks a little different this time. Your control panel is on the left hand side. Here, you can see our contact details the chat box and the handout session section that's circled in red here. Normally we would send you links to resources and other support material to supplement the presentation throughout the webinar. Now all the links and resources will be available in the handout section for you to view at any time.

Another feature I want to show you today is the chat box. Press the speech bubble icon in your control panel, and it will open. Click it again, and it will close. To talk to us, type a message here and click send these messages will be sent directly to us and we'll email you back a response within two business days. Now, I'd like to hand over to Steven to begin the presentation. Thanks Steven. 

SK: Again, thanks Brit. Welcome everybody again. This is a really nice session to have and well, I thought we'd start by talking about the learning outcomes that we should be all gaining by the end of the session. So we really want to have some understanding about high-quality patient health records and how they work to enhance patient care.

We want to have in our minds some ideas about strategies to improve the quality of the health records within our practice, and we'll talk about four specific types of clinical data that we can then modify in order to improve the record. They are of course not the only sub-sections, but we thought we'd focus on those today as the initial step. 

So the agenda today is really we will present a bit of the case for improving data in health records, we’ll talk about the attributes of high-quality health records, some strategies at the practice level for improving those health records, and again those for specific types of essential clinical data that we’ll try to get everybody to focus on a little bit.

So the question that that we're posing today is: why should we bother to improve our health record data? So we're professionals, health professionals. We have an aspiration of excellence to care for our patients, and as part of that health records should reproduce… replicate that level of care.

So the purpose of the clinical health record is to contain information about a patient that's required for their effective care. Good patient information supports appropriate clinical decisions. And that's what it's all about. It is having the correct information at the right time, at our fingertips in order to make appropriate clinical decisions for the best clinical outcome for our patients.

Other benefits, of course… good clinical records improve communication between health professionals. Clinical records will allow patients to understand their health care better when there's clarity and simplicity applied to the clinical record. The practice will be able to manage its recalls and reminders, and connect and communicate with patients, again for their benefit – for the patient's clinical benefit. Of course, the clinical record will allow us to more easily defend medical legal issues that we come across, and also have some visualization from a practice point of view about our patient cohort, about the entire patient group that we see within our practice.

Similarly, practice accreditation and quality improvement activities which are now linked to PIP payments (Practice Incentive Payments) will be easier and more simple to generate with improved clinical records. So once the clinical record is at a high level, the accreditation and the PIP payments system will be will naturally flow from that, and these are all guided by peer standards, by standards of expectation as decreed in the Standards documentation for published by the College, currently at the fifth edition.

So the attributes of a high of high quality records, and we need to break them up a little bit so that we can have a little bit of a finer, granular view of what we're trying to achieve with the with the clinical record. So, accuracy first of all, the our clinical record in general practice should be a major source of truth. It should have most of the information about the whole patient contained in it. So the expectation is that it's accurate, that it is in fact the source of truth; that it's complete, so that it reflects the entire patient as a whole, not just the surgical issue or the psychiatric issue, but the entire patient, and including a commentary about any individual condition. It should be consistent. So there should be a consistency within the wording, and within the way the phraseology that's used within the clinical notes that spans across the world. So we've all been taught medicine. We don't need to reinvent it. Our predecessors created a template for us to take a history, to examine the patients, to prepare some sort of management process. And that's quite consistent and that should be reflected in our clinical notes as well.

Legibility has now been taken out of the picture because of course we most of us are using clinical software. So, of course the clinical software will be legible because it's typed into a computer. However, spell checking is important, getting the correct information into the correct place and it's always preferable to use drop-downs rather than typing ourselves. We make spelling errors. The computer cannot understand a spelling error. So having that consistency and legibility and ‘understandability’ is valuable by using dropdowns.

Accessibility really speaks to the ability of somebody outside the practice to obtain our clinical record and be able to read it. And progressively there are some interoperability trials and challenges occurring at the moment. The CSIRO are involved in this. In order to be able to transfer records between practices in a more seamless fashion rather than the clunky way that we currently do it. So that will all evolve over the next few years.

And of course the clinical record of each patient needs to be up-to-date. So that really demands a constant curation of the record. The record needs to be updated. Every time we see a patient to improve it to make sure that it is in fact, correct in a contemporary fashion so that we're not missing any information.

Problems with the clinical record, and there are many, and most of the time the problems can be broken down into the data list that you see on your screen now. So it's often related to information being duplicated or missing or automatically put into the record from the computer system. So we need to be on top of that. We need to be on top of that and make sure that that is happening as little as possible, and when it does happen to rectify that process.

So we'll talk a little bit now about what we can do at a practice level in order to improve data quality, to get a universal buy-in across the practice to improve the levels of detailing and information that's going ahead. So there should be, inevitably, some sort of policy and checklist regarding the computer systems and the data entry that spans the practice so the practice management system, whether it's the owners or the practice manager, should create some sort of process where everybody is aware of their responsibilities when entering data, when deleting data, and that it's retained in a secure form that it can't be mishandled, and it's that it's protected for the patient's security and privacy. So these are very important things they roll off the tongue; they’re very easy to say but to actually make them happen can be quite challenging.

So when we educate our staff, we need to immediately engage in our computer systems and teach our staff precisely what is expected, that it's a whole of practice process that everybody will be involved, and we need to explain the reasoning why we do things. So why do we feel the need to make sure that everybody's address is correct, that everybody’s IHI is sitting within the clinical software, that we are only using drop-down lists and not using free type in the past history. In the commentary for a past history item – absolutely, the commentary requires, it demands, some free typing exposure, but the actual disease state should be a should be a drop-down. So there's a constant dialogue with all members of staff, both clinical and administrative, to enable them to feel confident that when they're using the clinical software, that they're not scared of it, that they're not going to muck it up, and that they in fact have a positive contribution to that patient's clinical record and hence ongoing care.

And all of those processes should be measured in order to grow, so that the practice grows, the practice advances, and we can use our data extraction tools. Many of you will have the PEN CAT or POLAR as extraction tools within your practice software, but the software itself will have ‘searchability’, so there are search engines within the software to look for things, and to therefore to use that software, to use that information in order to grow the practice and further teach every member of the practice how to enhance those that practice, that patient clinical note, and therefore expose all of the practice to the thread of the practice, to the management plans, and to the further development of the practice along the way.

And we'll use a case study or really a concept issue. So diabetes is a very common disorder. We know that many of the population have diabetes. So diabetes as a searchable item within our clinical software is available, so we can add a few clicks of a button we can identify how many patients have diabetes listed their past history. Now if we've got incomplete or poor medical records, some patients may have diabetes but may not have diabetes listed as a past history item. So we can search the database for people who have medications for diabetes without the diabetes diagnosis, and then rectify their clinical record. We can search the database for the people who have had an HbA1c (glycosylated haemoglobin) at all, or have it within a certain range, to identify those people who are well who are diagnosed with diabetes, or have what level of control of their diabetes, so we can then control, we can identify a group of patients who has, for example, poorly controlled diabetes. 

So let's say for instance, again, we'll use an example. Let's say for instance we would like to know how many patients in our practice have poorly controlled diabetes. So we will set our search engine up to search for people noted with diabetes, or with medications for diabetes, who have an hba1c in the old terminology more than, let's say, 8.5%, and we will generate a list. Hopefully that list isn't very long because I'd like to think that we all have lots of diabetic patients who are well controlled but there's often a few outliers. So those patients clearly having poorly controlled diabetes are at risk of all of the complications of diabetes, be they renal, cardiac, et cetera. So that gives us an opportunity to focus on that group of poorly controlled patients, in order to give them extra support, diabetes education, set them up with an exercise physiologist or other allied health practitioners, and therefore improve their clinical outcomes.

So we can only do that by being able to search and it's much easier to search in the computer system if the clinical record is as good as possible. So you can see that having a good clinical record will enable an effective search, and then will enable improved clinical outcomes. And that's only one example, of course, there are many, many examples. We haven't got time to discuss in any great detail, there’s so many examples, but each clinical state can really be examined and identified by searching the database in a similar fashion to therefore extract improved patient clinical outcomes once the issues are addressed.

So that then brings us to the concept of creating an action plan and allowing us to think of a condition or a problem that we would like to rectify and using those types of techniques of interrogating our database in order to get as much possible improvement clinically for our patients as possible. The PDSA process, that's ‘Plan, Do, Study, Act’, is a wonderful concept to use as a team leadership to use and to improve the quality of our care within our practices.

We really speak about a culture. This is a cultural change of not just seeing patients coming and going in our consulting rooms, but rather engaging in a whole-of-life care for our whole patient, the entire patient as one, and it goes right through the practice. So assessing the database is one component, then setting up targets for staff to aspire to, and all the staff are naturally competitive, especially the doctors, so setting up KPIs – that's Key Performance Indicators – within the practice so that they can then be reminded to improve their data entry, reminded to allow sufficient time to prepare and curate a patient's record will then create improved record outcomes which you can then measure again after a period of time and applaud and reward those particular clinicians who are performing. Well, it's a very good idea. It's a very good… setting up a solid positive culture within a practice.

So let's talk a little bit about which data to be involved with, and of course there are many data sets within our clinical notes. So for today's session we want to be a little bit targeted and just give some ideas and get your mind ticking, inspire you to actually move forward.

So we'll talk today about for specific types of clinical data as they're listed above and each one has its own little nuance. Each one has its own reasons – the why. Why do we want these particular issues to be addressed, and in order to create a better clinical record and improve patient outcomes. So let's talk first of all about allergies and adverse reactions.

So first of all, after some incentives, most practices have substantial, the majority of patients have an allergy recorded or identified that there are no allergies within their clinical software. So that's wonderful, but we need to improve that. So every single patient, there's no exception, every single patient should have either their allergies inquired about, and if there are none, then noted in the clinical software that there are no allergies. I think that should just be a blanket rule. And there's no reason not to – it’s one or two clicks on the computer and it's a very simple process to go through. 

So the allergies themselves… a lot of our colleagues in the clinical software write down the allergy according to the proprietary name of the medication, for instance. So I'll use an example of Augmentin. So quite often you see an allergy of Augmentin correctly into the file into the clinical record, but of course there are two components to Augmentin. So if the next doctor tries to prescribe Clamoxyl, which is a different brand of the same medication, the computer will allow this, by and large, more usually will allow it. So it is far superior to write down the generic drugs as allergies, because then when the check is made in the background by the clinical software you will identify Augmentin and Clamoxyl, because both contain amoxicillin and clavulanic acid, and neither will be allowed to be prescribed without an override from the from the user. So certainly the habit should be noting down allergies as generic medication rather than proprietary medication, rather than by their trade names.

The severity of the reaction and some sort of detail of the reaction: in most clinical softwares it's possible to put down a commentary on what actually happened in the allergy field, so you can have the allergy as a rash and then you can make notes about it underneath to give some qualification so that it's not just a rash. It's actually, you know, you're allowed to add information so value-add that particular recording. And so you can then on the next slide you'll see you can differentiate between an allergy and an adverse reaction. You can enter the severity of the reaction. If you're clear about it, you can then use that along the way and of course to get your practice up to speed as best as possible, you can certainly use an extraction tool so you can search your database to see who has any particular allergy or who has no allergy recorded, and take steps to rectify that to improve the clinical record.

The past history is always an interesting field to contribute to. Certainly there's a lot of redundant data in the past history and I refer really to multiple items, so often we find that hypertension is listed five or six times in a clinical record. It really only needs to be listed once. We often find that UTI or URTI so urinary tract infection or upper respiratory tract infection is listed many times. Prescription repeat or pathology results given are items which probably don't need to be in the past medical history. They don't actually add to the clinical flavour of the patient. And often that's because they're automatically put into the past history rather than manually done. So in your clinical software in your preferences, I would suggest you un-tick the box that automatically puts every history item into the past medical history. 

We should always use, as I said previously, we should always use drop-downs rather than free typing. We often when we're tired will reverse letters or make spelling errors of some sort, and that then comes up as an item that the computer can’t recognise, so that then escapes searching. So these are the sorts of habits that we need to develop to improve our clinical record to make them to make them functional and appropriate and more useful for the patient for their ongoing care.

So with the clinical conditions, and you can see on the screen now, using the drop-down of shoulder replacement in the time, and then you can see the commentary underneath to put down some free type, you know, the you might be perhaps the type of shoulder replacement, the surgeons name who performed the task or the hospital in which it occurred, whether there was any issues attached to that shoulder replacement, so it allows you to have some free type commentary in the bottom section there in the comments that will enhance the medical history and enhance the information as it transfers between practitioners.

The reason for visit is quite a rare subject item. So a lot of consultations occur and the reason for visit is not recorded. The rationale for putting this information in, first of all, there's a sequence and a logic applied to the clinical notes on the day. So that's the ‘SOAP’ that we've all been brought up with now, that the RACGP and the training schemes are teaching all of our registrars. So the subjective information, the objective information, the management and the formulation, and then the plan of what we're going to do with that patient. So as part of that, the reason for visit is a very big part of that. So what your current thinking is as to what's wrong with this particular patient. Now should the patient – let's say the reason for visit is applied to that, to a particular consultation, and the patient returns five years later, then it's very easy to look in the software to search for that particular reason, and that will show in the software and you can go back to those notes from five years ago in order to see what the thought process was, whether it's yours or whether it's another clinician. So it allows for that clinical handover and ongoing continual care of the patient, by having that data set filled in and as accurately as possible.

Within your software again, if you go to the to the preferences within the software and you can see that this one happens to be Best Practice, but it allows you to make a choice to prompt for the reason for visit, or to enforce the reason for visit. So it's just above of the red arrow is enforce entry for reason for visit, and then at the red arrow on the left is prompt. So generally I put the prompt in, but you can certainly have enforce, and that won't let you close the patient history until you've got a reason for visit established and located there. The prompt will request the reason for visit, but will allow you to close that particular history if you need to. Again a very good habit to get into is just to write the reason for visit. And you can see down the bottom of this page. You can see the lower red arrow the reason for visit window and you can see ‘always add to past history’. So that needs to be unticked. So that's where a lot of the duplicate and redundant data continues to present inside the past history, that automatically records that reason for visit as a past history item, whether it's already recorded in the past history or not. It will just repeat that. So that's where we get hypertension listed five or six times in the past history. That's where that can live, or anything else.

The reason for prescription is a very interesting information to record. So let's say for instance you prescribe an antihypertensive medication. The option will be there to record the reason for prescribing this medication. It will be in your drop-down list because it will come from your past history, and that should actually be notated, so it's a double click or however it's notated within your software. So the reason the reasons for doing it is that it then sits there as a lock. So the reason for prescribing amlodipine is for hypertension. It's not for something else. So there's a logic there, it allows openness and transparency of your medical record to the next doctor or person who's reading that particular medication. But that reason for prescribing also sits in the patient medication list that you can print from your software. So the patient medication list looks like… I’m sure many of you have contact with the Webster pack list from the pharmacy telling the patient precisely when to take their medication and why. So it’s the ‘and why’ that the reason for prescription sits in, so you'll find that , your software will be able to present and print a prescription list for the patient. So it's for the patient, so that they avoid confusion… a very functional item saves the patient, especially when you're changing medication. It saves the patient needing to ask a second time and hence reduces the confusion.

The reason for prescription also appears on their Shared Health Summary that you'll be uploading into the My Health Record. So that also is open for the patient's viewing. Of course, it's the patient's information, so that then allows the patient to utilise that information.

So there's lots of lots of things to modify there and it’s really a matter of using clinical relevance to change your habits when you're prescribing and when you're curating the medical record to improve the outcomes, to improve it, to make it in fact a lot better.

So in summary, we'd like to think that good patient care is directly linked to having high-quality health records. That's there's a number of papers have been published over the last ten or fifteen years, which directly correlate these two, to make the medical records accurate, to make them reasonable, intelligible, to make them consistent, to make them contemporary, so that they work for the patient and pass on the information. Of course, all of this information, well a lot of it, will appear in referral letters. So whether they’re referral letters to allied health or referral letters to specialists or referral letters to the hospitals, the referral letter will contain the past history. It will contain the medications and the allergies. So that sort of transfer of information, as well as into My Health Record, of course, all of this information needs to be up to date in order to further benefit the patient, and the practice can then use that information for development of its own protocols.

And we talked about how the whole of the practice can benefit with education programs, with developing the policies and creating an action plan to improve patient outcomes utilising the clinical software and the clinical notes.

The RACGP has produced a number of resources which will help you. They will help your practice to develop these ideas in a more formal sense and to bring that all home to improve patient outcomes by improving their clinical record. So the one on the left, Improving health record quality in general practice, was produced only 18 months or so ago, and really allows that understanding of what the clinical record can do and to teach ourselves and then to teach our practice to utilise that information as much as possible. 

Of course, the Standards for general practice as mentioned before, we're up to the fifth edition. Of course vitally important to utilise those standards in order to pass accreditation, but it's not just about accreditation. These are peer standards – this is what the community expects our practices to do and how our practices should perform. So it's not a matter of passing accreditation – accreditation should be easy, because you should be up to standard anyway.

Recently the PIP QI has been added to the PIP process where we've been incentivised to utilise our data in order to extract certain pieces of information. So the first step is what we're up to with the PIP QI and that is a leading process to then utilise our clinical softwar,e to utilse the data that we've that we've applied, in order to guide and support our medical practices.

The CSIRO are in the development phase, creating a data dictionary. So that will create some level of standard or at least guidelines for all of information to be on a similar base level,so that it can be compared. So we can compare a patient cohort in one part of the country to a patient cohort in another part of the country. It allows processing to work because we're using improved data of consistent framework, and we're waiting for further additions and further modifications of that process to occur, in order to enhance that and improve our patient and population health outcomes across the country.

BH: OK, thank you very much for that Steven. We do have a few more slides left and I just want to direct your attention to the RACGP’s Practice Essentials Webinar Series that we're running for the rest of the year. This is a monthly webinar offered by the RACGP’s Practice Technology and Management Unit. In 2019, this was offered as the eHealth Webinar Series. Upcoming topics include privacy fundamentals in April, medicine safety in May, and owning and managing a general practice in June. For more information and to sign up for future sessions, you can visit the RACGP website.

Now we do get some questions on the topic of data quality that come up frequently with our team. So I'm just going to ask a few of you now, Steven. The first one is: can patients have access to their health records?

SK: OK. Thanks Brit. This is a really common question that gets asked, especially with computerised records. So the traditional thinking was that the records are owned by the medical practitioner, so the author of the tecord was the owner of the record. With Freedom of Information and the modern generation, that probably doesn't really hold on anymore.

So realistically the record is sort of owned by the patient. So if the patient wants desperately wants to get their medical record through FOI, through Freedom of Information, that can be made available to them. So it's certainly challenges us as the authors of the record to make the records logical and sensible and consistent throughout. Of course the record in its entirety sits in our general practices. As I said before, it can be accessed through legal means, through FOI. Sometimes there is there are legal requests for the medical record. Within my health record of course, the Shared Health Summary gets uploaded, which is only the four areas of medication, allergies, past history and immunisations. Those four areas as shown within a Shared Health Summary, and depending on how you write your referral letters, various amounts of information will appear inside referral letters sent either via paper or via secure messaging delivery (SMD). The most usual way is HealthLink or Argus. So the information that we produce can be seen in specific arenas, usually not by the patient, let's be fair, it’s only rare that the patient actually wants to see their record. But even so, the data quality, the quality of the record, the style in which we write, the accuracy of the information all needs to be the highest level and highest calibre to secure appropriate information transfer to wherever it's going to be received.

BH: Great and we have time for just one more question. Steven, what is the best way to keep notes brief to save time, yet satisfy the legal requirements and still maintain good quality health records?

SK: So I guess the we're in this transition phase between paper records and electronic records. So most practitioners around the country have never been taught to touch-type. Certainly the younger generation, and we're seeing it now with registrars and junior Fellows are certainly able to touch type, and their typing skills – the speed at which they type with tremendous accuracy – is really something else. So there's an onus on those who can't touch-type, or those who don't type very quickly, to actually upskill to go and learn how to type better. So it's the actual typing. It's when we were in school, we learned how to write. Now we need to learn how to type, and it doesn't really come naturally, it needs to be practised. So that's one component to improve the speed at which we consult and record our notes. 

The other thing is, of course, to use the computers much as possible. So to use the drop-downs, to use the radio buttons when they're appropriate, and to use those and then nurture those to get the information across, but often radio buttons or drop-downs will give a lot of information within the clinical notes with very little effort. 

And the third thing is really to improve the auto-fills. So shortcuts of various sorts can be retained, and then stored in the auto-fill section within your clinical software. So you just need to type in a code, which will then write notes into your clinical record. For example, I have a code called ‘sunn’, which s-u-n-n, so it's not a real word. I've put an extra ‘n’ at the end of the word. If I type in ‘sunn’, I have a three or four line description of the discussion that I've had with my patient about sun avoidance, about sunscreen usage and about care afterwards with moisturiser et cetera. So that enables a protracted discussion to be reproduced within the clinical record with four key strokes, which is s-u-n-n, which is very, very simple. I have another one called ‘RTI’ – respiratory tract infection – which gives a framework history and examination which gets immediately dumped into my clinical record, into the notes for that particular day, for that particular patient, that I then adjust. So it says ‘sputum’ and I say ‘no sputum’ or ‘sputum green’. So I modify, I curate, the auto-fill message in order to I will only write one or two words, knowing that the template puts in the framework of what I'm going to say, and it becomes quite accurate and meaningful as part of that. So those three things will help you to type quicker, be smarter with drop-downs and the way things get stored and reproduced within the clinical record, and to use the auto-fill in order to save time and make your records more accurate and more appropriate.

BH: Great. Thanks very much for that Steven, and thank you for a very interesting presentation on an important topic.

If you have a question for us that you haven't already entered in the chat box, you can email the RACGP’s Practice Technology and Management Unit at ehealth@racgp.org.au, and the team will get back to you within two business days. We hope you found this webinar informative. Thanks for attending and hope you'll join us for a future webinar in our Practice Essentials Webinar Series. Goodbye.