Saturday, 13 December 2008

Non Drug Allergies

Non-Drug Allergy Checking

This from Laurie Slater on the forum (and gp-uk) regarding Nut Allergy and Abidec:

"...although nut allergy was read coded in the history, there were no
flags raised on prescribing this medication.."

But AFAICT it should work and have tested this morning as below.

Add code SN582:

then Vision opens the Non Drug SDA:

So 'OK' that then try and prescribe 'Abidec oral drops':

The Peanut Allergy does not appear in the red warnings but, nevertheless, when you click OK:

So, it does work if recorded in the SDA. I wonder if it was recorded as a Medical History item?


Read more...

Monday, 24 November 2008

desktop background

Ah well, this can be changed but you need to go digging in the registry, which I would clearly never do.

Also, if you then set the wallpaper through Firefox or MSIE then it will work as the control panel interface remains disabled via the group policy. Again, a registry key controls whether you see the wallpaper tab in display properties.

It is useful to have different wallpapers for different users as this then is a quick way of seeing if you are the logged in user as your desktop is immediately identifiable.

Not sure if changes I make locally will persist on reboot and relogin though. Find out tomorrow.

Read more...

new server

Hey ho, a new server at the weekend. DXS is broken. Foxmarks vanished (but I have fixed this - you need to tell it to use encryption and it works!). Can't change the wallpaper. This may simply be access to control panel that has been disabled..I feel challenged to work around it, even though am normally content not to have fancy backgrounds etc. I get the feeling that Clear Type is off by default. That is annoying.

This is because the server, and thus our logins, have been moved to the Glasgow domain, which is all locked down. Don't know yet *how* locked down it is, but will post any gripes as I discover them.

Annoying DXS being broken though...

Read more...

Thursday, 20 November 2008

post conference

Well, that was a good conference. Well attended (despite the economic climate) and useful talks. Thanks to anyone who attended mine and patiently put up with me talking through a cold which still refuses to go away...Thanks also to the organising team!

Reviewing the prescribing safety settings for NVUG it is pretty clear that in all parts of the UK bar Scotland at present you are able to tweak these pretty much back to where they were prior to DLM 260. We north of the border will have to be patient till we see some relaxation of the rules.

Nice stuff coming up in Vision in the next few DLMs and I am looking forward to both the Free Text search facility and the changes to the problem management. I have largely given up on problems for the time being, but will revisit them when the changes are delivered.

SNOMED talk was also fun, but apologies to anyone who thought it might make some sense! Is SNOMED really fit for purpose? I guess the answer depends on what the purpose will be. One of the biggest difficulties with SNOMED is the lack of people in healthcare and developer communities who understand it. It is complex, perhaps necessarily so, but does the inherent complexity of it make it essentially unusable in the real world? I think we need to get end users educated in SNOMED as far as we can, as the end users can really only drive the functionality if they can understand (to an extent) the advantages and limitations of the terminology.

Anyway, feeling fairly crap with this cold and have work to do...

Read more...

Monday, 10 November 2008

Conference

Well, the new NVUG web site is up and running so well done to Simon Child for this. And welcome to the blog if you have navigated from NVUG.org to find it.

Conference this week so I am busy (overwhelmed, perhaps!) writing talks for the 2 days. Interesting breaking down the decision support for the 'Prescribing Safely' talk, and hope is that in Scotland DLM270 will return us to something more usable and safe. The SCIMP conference talk on Prescribing Safety was also enlightening, particularly the input from FDB on the difficulties that computers have working out if a patient has a condition.

Always reassuring, this, as it means GPs cannot (yet) be replaced by machines. Ahem. :-D

Also talking on SNOMED at the NVUG conference, which will be a) a challenge and b)slightly scary. I am expecting two people to turn up, right enough, one of whom will be lost and the other probably a clinical terminologist for a living. Joking aside, the challenge of moving the NHS people to using SNOMED is huge. Read has suceeded in part because people can understand it with only a little help. Thus they are able to explain it, and use it with success, in normal work. SNOMED is not easy to explain, nor immediately easy to use. I guess we are relying on the systems implementers to make SNOMED work under the bonnet, whilst all we need to do is steer in the right direction. Can that work? Let's hope...

Read more...

Tuesday, 21 October 2008

DLM 260

Ah...we got this. Med 3s in an SDA, need to make a decision as a practice if we use these or just use medical history. Mucks up our 'certificates' guideline also.

Immediately wrote a macro to do 'yes' and 'proceed' for drug warnings. I was used to working with the warnings previously, and now find I cannot easily filter out the stuff I want to see. Humph. Soon to be fixed in Scotland though, I believe.

Read more...

Problems

Worked with problems for about a month now and variable experience with them. Perhaps I need some training, not so much in the how to use the software, but perhpas more in how to run and operate a problem orientated record.

There is an overhead in maintaining the problem list, and making sure stuff you enter goes under the correct problem heading. Some of the V3 tools make this relatively easy, but overall it is a footer to maintain. If I get the chance I will try and attend the problems talk at NVUG this year, which may help.

Certainly it is useful to see a list of active problems for the patient when they come in the door, and makes it much easier to sort and recall what the current issues are.

I'll keep going with it for now. Getting the rest of the practice team to use them effectively would seem quite a difficult task though.

Read more...

Monday, 22 September 2008

Problems

Panel meeting at the week end mostly sorting out the conference.

I am going to experiment with moving to Problems, given that a couple of other panel members have been using these with a degree of success. I have always been put off previously by the data hygiene overhead, but we are a small practice relatively IT capable so maybe we could make it work?

Download the problems section of the training manual from here.

Read more...

Friday, 19 September 2008

How to search

OK, so not been so productive on the blog of late. Sorry. Anyone is welcome to join as an editor though, so let me know and I will sign you up (Karen!).

Here is how to do basic searches. If you have not explored searching, or were well turned off it by GPASS functions in this area, then it is a worthwhile exercise. Honest. It is easy to create a simple yet useful search and, if you wish to explore further, Vision 3 searching is flexible and powerful for more complex work. It can become confusing when you combine a search with complex reporting logic, but I am only touching briefly on that today.

OK, open Searches and Reports from the Vision 3 main Screen. It's under 'Reporting':



And this opens (in Glenburn) this:



which is immediately confusing! Essentially, though, it is quite feature rich and if you have not looked at searching previously it can seem a little daunting. Do not panic (at least, not over this). I am only going to demonstrate how to create a simple search, so click on the button as below titled 'New ad hoc search'.

OK, this opens the main 'New Search' screen, where you create new searches. This is where most of the work in searches is done.


You will see the form is divided into useful sections.

Group Input / Group Output
Input allows you to choose a group of patients that has previously been saved and use them as the input into the search. You get get such a group by running a search and saving the found patients as a group (Group output) or but saving the group from clinical audit.

Search details
This is where the meat of the search is configured, and is discussed in detail later.

Report Output
Allows you to configure how the search results are presented. The default is 'View' on screen, but other options will present you various amounts of printable detail.

Add Entity
Click the 'Add Entity' button next to 'Search Details' and you see this:



which may or may not be familiar to you. This form allows you to select from any of the Vision 3 'entities' for searching on and, once selected, you are able to search on the attributes of the entity....

Yes...?

Ok, well an entity is a thing that you might have recorded in the patient record, like a 'Blood Pressure' or a 'Weight'. And 'entities' have 'attributes', i.e. stuff around them that you record so 'Blood Pressure' has attributes of 'Date', 'Diastolic', 'Systolic' etc. So really, you add things to search on and then tell the system what values of that thing you are looking for.

Here we navigate to 'Blood Pressure' and OK that:



And 'OK' brings you back to this:



For our 'Search Details' we want to just search on our current active patients. Vision 3 keeps all your transferred out patients on the system (as do other clinical systems AFAIK) and, by default, searches are run on every patient in your database. To tell it to only search on our current patients we need to click on 'Patient Details' and then 'Selections' and tell V3 to look only at our 'Registered' and 'Applied' patients:


This opens the 'Criteria Select' form:


Navigate to 'Registration Status' and select 'Equals', then 'Permanent' also 'Applied':


OK that:



so we are making some progress.

We will set this up now to search on patients who have ever had a diastolic BP of > 85.

Select Blood Pressure in the Details screen and click on 'Selections':


Similar from to the one for Patient Details, but different criteria of course!

Navigate to 'Diastolic Pressure' and select 'Greater' and check on 'Inclusive' and add the value '85', so really we are doing a 'greater than or equal to 85' search.:



Ok that, and here is how it looks:


I only want a count of them, so change the 'Report Output' from 'View' to 'Count' as below:


Our search now looks like this:



Click on 'Run' and wait a minute. The search estimate times are always much longer than the search actually takes. I think this is just to make you feel better...

Here is the result:


So that is 935 patients who have ever had a diastolic blood pressure of over 85. They have 4037 matching BP records in total.

Hmm....be more useful if we could find the patients whose last bp was >85. OK, every time I visit this I get confused and it does take a little bit of brain challenging logic. To do this, (R) click on the Report Output 'Blood Pressure' and select 'Options':


This form then opens. Read it through, then read it again...

It does make sense, but you probably need to read it every time you come here. The option we want is 'Before Match'. In other words, after you have found all the blood pressures but before you report on this, check only the last one and only include it if it meets the criteria as above.

So now, when we search we get:


So, 540 patients last BP was > 85. That seems a lot.... better get back to work and stop this blogging malarkey.

Save the search before you exit. Call it something sensible, and in the description prefix it with something (I use my initials) so you can find it again later. Ad hoc searches sort by the Description in the list, not by the name.


OK? Brief searching how to.

NVUG panel meeting in Birmingham tomorrow. Conference planning amongst other things. I am in Dundee currently, though, so a long way to go!

Read more...

Monday, 25 August 2008

Drug check in F3

When searching for a drug if you (R) click on it you can select 'Drug Check'. like so:



which, usefully, will run decision support checks against the drug before you have selected it.

Read more...

Thursday, 7 August 2008

Dealing with journal clutter

The Journal list is a list of everything that has happened for that patient and, consequently, trying to use it for day to day stuff can be troublesome as it fills up with repeat issues and pathology. Cutting through all that to see just the narrative can be done by setting your 'initial filter' to exclude those items.

Right click on the 'Initial Filter' on the Navigation Pane and select 'Show items with no data'. You have to do this so you can select from all the possible items on the list to exclude or include them.


This done, you then go through the list and, holding down CTRL, left click to select items.



Then (R) click on the 'Initial Filter' again and select 'Save Selection as Initial Filter'.



You have to make sure you have a Filtered Journal List in your tabs in Consultation Manager for this to work. I have this set as my default tab on entry to the record, but I also keep the full Journal view in a tab adjacent to it.

Read more...

Thursday, 24 July 2008

New Computers

PCT (or whatever they are called nowadays) has given us new kit :D. Lovely HP P2015 Laserjets and dual core Pentium Dells.

Nice.

Had to take my 23" Viewsonic home and swap for the 21" as for some reason the graphics in the new PC does not support the resolution needed by the monitor. The smaller monitor does not swivel, making harder to use in the consultation but at least I can read what is on the screen! The graphics card for some reason supports resolutions below and above the required one, just not the actual one I needed.. :-(

Read more...

More SEF

Back from Tuscany and only now recovering from the shock of returning to work. Tuscany fab. So now I have calmed down and had some time to investigate further it is apparent that this SEF Prescribing Safety problem is a much more complicated issue than it first appears.

First off, it is clear that the intentions are very sound - make our systems' method of drug alerting more in keeping with the requirements of our clinical practice. What is not yet clear is why the impact of this has been so negative.

End result anyway is that we have a system which was usable and reasonable (albeit with some flaws and omissions) with respect to drug decision support and now is dangerous and frankly irritating! I see in England INPS have prepared DLM 261 to deal with the issue, but here we are reliant on SEF approving a fix of some kind before we can move forward. That is OK, provided it is not too delayed. There are a lot of positives in the prescribing safety recommendations, and it would be nice not to lose them.

So I guess in reference to my last rather irritated post it would be fairer to reflect that there is no easy fix here, nor any easy way to apportion blame. This is difficult stuff to do and there are now a lot of people working on it. Let's hope we get something decent out of it in the next few weeks.

Read more...

Thursday, 26 June 2008

DLM260 and Prescribing Safety

This DLM contains a lot of stuff. We do not yet have it in Glenburn, but given the upset I am not so sure I am looking forward to it!

The 'Prescribing Safety' implementation aspect has come from the Scottish Enhanced Functionality programme - essentially a method the Scottish NHS employs to ensure certain requirements are met by system suppliers or else the Scottish NHS will not pay for their systems. The main aspects of the Prescribing Safety SEF that have caused controversy amongst users are the appearance of a new dialogue asking for a 'Reason for Override' essentially, this against all 'High' level warnings. This is probably dangerous as it risks alert fatigue.

There are some other aspects of the DLM which have led to confusion so I will attempt here to explain medication decision support options to let you, the happy users, tweak the settings for your continued enjoyment and benefit. ;-)

ALT COS

Before we go any further, if you have never discovered the Consultation Manager, Options, Setup screen (accessed with ALT - C - O -S from the keyboard) then now is the time.
Which gets:
And you want to click on the 'Drug Check' tab which, for me (because we have not got DLM 260 yet) looks like this:
But with DLM 260 it is this:
The above screen lifted from the DLM 260 help file, which is available from here.

What has Changed?

Minimum Number of Days for Drug Check

There used to be two options here, as you can see from the original Drug Check form above. The first was for which Clinical Data to use for therapy decision support ('drug check'), and the second for which therapy records to use. They have removed the 'Clinical Data for Drug Checks' option entirely. Previously you could use an option of 'Currently Relevant' clinical data, but I am not sure (and few people seem to be!) what this meant. Now the system will check on all medical history, regardless of whether it is flagged in the system in some way as 'active'. (The only way I can see to flag a medical history item as 'active' is by using problems).

This change may have impacted on what you now see in your drug check warnings, as the system will now be including clinical data in its therapy decision support that it was not doing previously. To be clear, the clinical data checks apply only to Contraindications, Precautions and Prescriber Warnings - this is usefully referred to as 'Condition Checking' in FDB documentation. We have had some e-mail correspondence with clinicians affected by this. This screen shot:
was supplied, the patient being a fit young man. Why the warning? Well, because now ALL medical history is being queried and this man must have something in his history that is triggering the alert. But how do you find that out?? Click on the warning line itself, and it will expand to show you the triggering history.
Now you and I might think that 'Renal Calculus' is a different thing from 'Severe Renal Impairment' and I am sure that people at INPS think this also. I am also fairly sure that First Data Bank (FDB) who supply the drug dictionary and decision support think this is not a great alert either, but right now this is what it does. The other change here is the 'Therapy Data' for drug checks which has had the 'Currently Relevant' option removed and a default time period added. The help file says this:
'We recommend that all users set the date offset to 1 year. On receipt of DLM 260, all users who have an offset date of less than 30 days will be automatically changed to the minimum 30 days offset. For new users, the default Therapy drug check offset date is set to 1 year.'
In Vision 4 I am hoping to use 'Current Medication' as the driver for decision support, because 'Current Medication' is better modelled in V4 than V3 partly due to the different implementation of medication types. For the time being, however, in V3 (and V4) we are using a date offset. So for interactions and drug doubling warnings, Vision 3 will look back at the 'Date' of therapy items and check your newly added item against all items that come within that date range. I do not know at present which 'Date' it looks at, whether that is Date Prescribed, Date Issued, Date Authorised, but hopefully it is taking into account all 3. This screenshot:
shows an alert triggered for drug doubling although the patient last had a co-codamol prescription issued 4 months ago. This is doing what it should, and it is querying all therapy data in the past 1 year for checking for interactions and drug doubling. Again, you can click on the warning line to show the triggering medications:

Contraindications, Precautions, Prescriber Warnings

You can no longer turn these off, and if you had them off then Vision DLM260 will turn them back on for you at the level of 'Patient Specific'. 'Patient Specific' means the system will check all medical history for the patient and see if there are any relevant warnings, like the 'Renal Calculus' one above. The specificity of these warnings is being worked on by FDB (we hope!) so one day soon it will be better. Deciding if a patient has a thing is not as easy as you would think, in computable terms at any rate! You can set these to 'General' and 'all' also.
though I remain unclear as to what a 'general' warning is and still need to see if I can identify in FDB what exactly is going on there. I set mine to 'Patient Specific' and this is fine for me.

Interactions

They have taken away the option of turning off interaction warnings completely so you must see some of them. In Scotland you cannot only show 'High' warnings in interactions, you must show 'High and Medium'. These are actually level 3 and 4 FDB warnings.
"For Scottish practices, you can select High/Medium/Low, or High/Medium, to display drug to drug interactions, but you cannot select High."
You can in England, you lucky people :) So now we get all the 2 and 3 bar FDB Interaction warnings all the time, and cannot stop them. This is probably not so good, I think.

Drug Doubling

Drug doubling also has had the 'Suppress all levels' option removed, which is OK. You can still set it to Same Action group which is what most clinicians use IME (when they know it is there!).

The Override Confirmation Required Problem

So this setting makes Vision 3 give you an additional screen before being able to add the drug. In Scotland (again, for we are truly blessed) we are not allowed to turn this down to only High warnings. Which I think is the biggest safety risk of all.

We now have logging of overrides, so presumably if the prescriber duly gets into the habit of doing ALT+P for bypassing the warning reason (as they will have to to make the system usable), this is audited. And if it harms a patient, the clinician will carry the responsibility for this but NHS Scotland can state 'We made sure they saw the warnings, so it ain't our fault'.

The paradoxical nature of presenting every warning to busy clinicians who work using heuristic thought processes is that over warning them is dangerous, because it is only when something breaks the pattern that it stands out.

So, I would think that Level 4 FDB warnings and Allergies should require a reason and that should be the interruption to your usual, every day prescribing process that stops you killing folk. This is (almost) what you can set it up for in England. But not in Scotland. So, what shall we do here? I think this is a question we need to take back to NSS / SEF / INPS / FDBE and probably other vendors.

Anyway, the point I am trying to get across to those of you in England is that you do not need to see this 'Add reason' form every time because you can change it in Setup to 'Only High'. Which is quite good, really. You can bypass the override reason with ALT+P or clicking on Proceed. But I think getting into the habit of hitting ALT+P is the way to go. If it is just for high level warnings then entering a reason might even be clinically prudent!

I would be pleased to have Scottish user's comments on the Prescribing Safety SEF (e-mail me), whether good or bad, and I will feed them back via SCIMP to NSS.

More

There are other aspect to Prescribing Safety that I am not going to discuss here just now, largely because I have to get some paperwork done and have a surgery to do! Download the Help file, it is pretty comprehensive. I am away on holiday for the next 2 weeks so no post to the blog. I am always hopeful that someone else might want to contribute to the site, so please contact me if so and I'll set you up access.

Read more...

Thursday, 12 June 2008

Consultation Management

This form:



lets you define the attributes of the consultation. If you change the date here then every entry you make will carry that date. Similarly for the selection of a clinician.

Most importantly it also allows you to change and select an appropriate consultation type. This can be done quickly with the keyboard by pressing, repeatedly id needs be, the first letter of the Consultation Type.

Thus, 'S' for 'Surgery Consultation' or 'M, M, M' for 'Medicine Management'. It is good practice to use these types as best you can.

If you don;t want to start a Consultation with the patient, you just want a look at their records, then you can click on 'Cancel' here, which will display the record but not record a consultation against it. Pressing 'Esc' on the keyboard does the same thing.

You decide whether the Consultation Form appears at the end or the beginning of a consultation in Con Man Setup here:


So you can have it on open, close, both or not at all. Which is a matter of choice really, some find it easier at the start and some at the end. Unless you are doing exactly the same task every day with the record, I would recommend you have it appear at least once.

You can also bring the form back up to Update it by double clicking on the status bar on the bottom right of the Patient Record view here:



so if you need to change anything, you can!

Read more...

Tuesday, 3 June 2008

RHS Advice

Using the Right Hand Side of the prescription for an additional text advice is very useful but has no expiry date, which means it persists for ever more. I always start such advice notes with the Date on which they were added. Then, at least, it becomes apparant that it is out of date or no longer applies when you see it appearing 6 prescriptions later or, worse, copied to the repeat master text.

Expiry date on advice notes would be a good thing IMHO.

Read more...

Thursday, 29 May 2008

Drug Defaults

If you don't have these on then every time you attempt to prescribe a commonly used item such as Paracetamol tablets 500mg you will have to amend the default quantity (it defaults to '8' from Normalex) and probably the dosage instructions as well.

So: sort out drug defaults. You (R) click on the drug name and select 'Drug Defaults' -> 'Maintain' to do this.

Agree the usual defaults with your partners and prescribers to avoid arguments. Drug defaults will apply across your practice - not just to you.

If you start setting these up in an opportunistic way then within a week you will have the 10 common ones and by 6 months you will have 80% of them done. And it saves so much time and hassle it is definitely worth the investment.

If you cannot remember Paracetamol 120mg / 5ml dosages per age group (and after 14 years of GP working I still have to check them!) then setting these up in drug defaults is a little bit of magic.

Read more...

Tuesday, 27 May 2008

Prescribing

I went to Campbeltown to do some training work on consulting with computers and Vision 3. Thanks to all the team there, but this drew to my attention to how many practices never quite get around to setting up Vision 3 therapy correctly and this just makes your life difficult!

The steps to succesfully using Vision 3 therapy functions are to:

  • Setup a Drug Formulary
  • Set users to select from formulary
  • Spend the time setting sensible and agreed Drug Defaults
  • Understand what Dosage Codes do and how to ignore them

The first is essential as it limits your initial choice of drug. Remember: a drug formulary is practice wide, not per user, and you can only have one formulary for each practice. In V4 land we are going to extend this functionality for multiple formularies per practice but, as you can imagine, this may get rather more complicated!

You can setup a drug formulary by hand, by trawling the action groups and double clicking to select or de-select drugs or you can use the Drug Dictionary Utilities to quickly create a rough and ready formulary which you can fine tune later.

Enable this per user in Control Panel, Security, Edit User, 'Select Drugs from Drug Formulary' and 'By Drug Name'.

Then you are no longer selecting from all the drugs in the dictionary but from a subset, thus 5 paracetamol preparations rather than 100!

Read more...

Thursday, 15 May 2008

Drugs Clinic Guideline

Version 1.0. Designed in part to replicate the data collection required for reporting purposes.

Read more...

Tuesday, 29 April 2008

Holiday

Sometimes you just need a break. Back now though.

We have discovered Smart Priority Numbers for Vision 3 from Bradley Sieve. This is just the utility we need to sort out our mess of priorities, with our odd mix of 1s, 2s, 3s and more.

Not had the chance to play with it much as yet, but here is the main screen:



from http://www.tailormadeit.co.uk.

Clever :)

Read more...

Monday, 7 April 2008

With Macro Express

With Macro Express you can achieve the same thing more efficiently with respect to changing priorities.

This is the code:

Variable Set Decimal %D1% from Prompt
Variable Set Decimal %D2% from Prompt
Repeat Until %D1% = 0
Text Type: ey%D2%k
Variable Modify Decimal: %D1% = %D1% - 1
Delay 50 Milliseconds
Repeat End

and it will prompt you for how many items to change and to which priority.

Read more...

Thursday, 3 April 2008

Priorities .... again...

So we have adjusted our priorities to be 1 and 2 for high and 3 for routine, and that is it. Unfortunately we have a lot of Priority 2 items that are really routine entries, as '2' was default for medical history items for some time. Consequently, my Medical Summary guidelines has had to use P1 items then P2 items of type 'Diagnosis' as below:



In Edit mode:

But this means we have a lot of data tidying to do over the next year (or more!). Data tidying is similar to painting the Forth Rail Bridge, but less risky.

So, how do you change 30 P2 items to P3?

Keyboard Express, of course. :-)

Get this from:

http://www.keyboardexpress.com/

This is a keyboard macro program, that can play back a series of keystrokes automatically and efficiently.

Having installed this I created a macro with this code:

<REPEAT:0038><SPACE><APP>e<ALT>y3<TAB><TAB><TAB><TAB><ENTER><DELAY:00.20><ENDREP>

which will change 38 items (REPEAT:0038) to Priority 3 (y3) from a filtered journal list view in Consultation Manager.

So now I can filter the view in CM to show the P2 items I wish changed, note from the Navigation Pane how many there are and adjust the number in Keyboard Express accordingly. Highlight the first one and then initiate the macro and in aout 20 seconds 30 + items go from P2 to P3.

It would be even more useful if this would work on a Guideline view, but I think I would need to use Macro Express, which captures mouse movements and clicks to achieve this.

Read more...

Tuesday, 25 March 2008

Rec9

Doing the anlysis for this QOF indicator this week and last. We do not use problem linkage to our repeats to meet this indicator - I have always found it too cumbersome. Although possibly once the overhead of adding problems for the majority of them has been done it would be OK to work with thereafter.

Anyway, we use an indication in the Dosage field. For example, "Take two puffs twice a day [ for asthma prevention ]".

We routinely, as far as possible, use the square brackets to enclose this 'indication' text and we have set up most of our standard drug defaults to include it so that it populates automatically. Clearly this needs tweaking from time to time, as not all drugs are taken for the same purpose.

When it comes to analysis of this, I run a search and export the information as dbf tables, then import it into Excel and do some simple counts on the data to produce a percentage score. By this method we currently get 79.6%! Damn! Although this method includes all appliances and monitoring equipment, which I understand can reasonably be excluded from the QOF target. I have been unable in the search to exclude the Action group for Diabetes monitoring agents - it produces an error which sets the field value to 'null'. This be a bug, methinks.

The search is this:


and yes, I know, you can do an initial patient list search and then load that group in.

V3 does not know when repeats were commenced. It knows when they were re-authorised, but not when they were started as every re-authorisation event stops (inactivates) the previous repeat master and creates a new one. The QOF requirement only applies to repeats started after 1/4/04, but in V3 terms this means anything re-authorised after this time. A 'date commenced' is intended as a separate attribute for repeat masters in V4.

While we are on searches, if you save your search using a description prefixed with some initials, then it makes them easier to locate in the adhoc search list. e.g.:



So then I import the tables into excel and I get this:


The formula for counting how many have the square brackets is:

=COUNTIF(D2:D4358,"*[*]*")

79.69%

damn

Still, be fixed by end of month and really we are achieving this already, but it would be nice and reassuring to have a spreadsheet that clearly shows it.

Read more...