The Drug Intervention Log
Dr Ben Green, MB, ChB, MRCPsych,
Consultant Psychiatrist and Honorary Senior Lecturer, University of Liverpool, UK
What is microaudit?
An audit of a service or the practice of a group of individuals in terms of product or
process can be termed macroaudit.
Microaudit attempts to formalise the internal auditing practices that all professionals use to calibrate their performance. When doctors decisions regarding clinical practice they often use their own subjective experiences of cases they have seen, and their responses to treatment.. Sometimes the use of these 'subjective' experiences or anecdotal practice is decried. This is very much the age of 'evidence-based medicine' after all. The internal calibrative processes that professionals tend to use (ie their experience') is denied the value it once had. This is all very well, because after all, instainctive cures and theories based on bigotry, or authoritarian opinion rather than authoritative fact, was very much the state of medical practice in the nineteenth century and earlier.
Scientific method has yielded great research findings. Although microaudit should not bee seen as a research exercise it does tend to objectify the sometimes distorted subjective recollection of the good some action has achieved.
Microaudit in practice
There is no government quite as satisfying as self-government. In the UK audit was very much imposed from 'on high' as a result of government policy and legislation. Nothing rankles so much as something which has been imposed from above (even if there is a good idea behind the imposition). The philosophy of microaudit is therefore self-government. If the process is itself fulfilling and rewarding then the process will be repeated.
As an example of microaudit let us use an example looking at the continuous process of prescribing drugs for patients seen in a clinic or other clinical environment.
To audit the prescribing habits of an individual requires some
standard instrument or tool. In this case I devised the Drug Intervention Log, an example
of a microaudit tool. The microaudit tool needs to:
- be simple
- have face validity
- be quick to use
- be adapted to the clinical situation
- be easy and quick to analyse
- provide meaningful information
The Drug Intervention Log
The majority of clinics and practice environments do not have elaborate prescribing software available on every desk. The log could be easily integrated into the prescribing process, but can be quite conveniently used on paper.
The log records every pharmacological intervention from a certain time forwards. There is a baseline recording of the drug regime at the initial log session. Changes may include:
- increasing the dose
- decreasing the dose
- changing the frequency of the dose
- changing a drug
- stopping a drug
In each of these cases there is a reason for the intervention (past or present) and the doctor has some intent in mind (a hypothesis really). The log records the change, the reason for the change and the doctor's anticipated intent. Formulating what the intent is precisely enables the doctor to know whether the outcome of the intervention is successful. It sounds simple and rather too obvious to attract much attention or brainpower - but notes all too often do not record why drugs are changed, or even what the change was. Clinicians are familiar with the experience of poring through paper notes, deciphering others' handwriting, and generally trying to understand long-forgotten, and often very clinically important events.
Used responsibly the drug intervention log can provide a historical perspective and reinforce rational prescribing.
An Example using the Log
Below is a figure of the first entry into a Log. The patient, a 33 year old woman with a biological type depression, has been on a short course of sleeping tablets for insomnia. The doctor elicits features of major depression in the interview and decides to prescribe an antidepressant. He hopes to see improvements in her sleep, her mood and her appetite. This is his intent, and later he sees the patient in follow-up situation to record (in absolute terms) whether the hypothesis that he formed (that an antidepressant will improve these features) is correct. The patient's sleep has improved, and so has her mood and appetite.
This is not a rating scale. Its measurement is not designed to be so precise. It can give the doctor and the patient feedback on the benefits (or otherwise) of prescribing certain drugs in certain doses in certain conditions.
The prescriber him/herself sets the targets that the intervention hopes to achieve - allowing the prescriber to calibrate in a much more formalised way. The self-feedback is as objective as the prescriber wishes it to be.
Beyond the Single Intervention
Multiple interventions may be necessary in some patients eg to control blood pressure in hypertension or fine tune depression. The log can record this approximation process. Interindividual variability in the response of different patients to the same drugs can be judged better. The clinician can use it to formalise an audit of his/her practice - is their practice of starting patients on a low dose of depot antipsychotic really a wise one when multiple patients treated in this way in their practice show a regrettably slow response?
The DIL factor
The DIL factor is simple enough to calculate. This is the percentage of drug interventions deemed successful. It is probable that different specialities can expect different DIL factors. In anaesthesia if one were to log all anaesthetic inductions against the target intention of inducing unconsciousness you would expect the percentage of successful interventions to be nearly 100%. (Although anaesthetists could fine-tune the DIL to look for other outcome criteria eg absence of cognitive disturbance post-operatively).
Just as macroaudit has its audit cycle - [based I believe on the PDCA cycle introduced into the Japanese factories by the Americans after World War Two (What's PDCA ? - perhaps we'll leave that for another time)]. - so does microaudit. Imagine finding oneself with a DIL factor of 30% - a sort of placebo doctor - you would expect that further training might see a rise in the DIL factor. Hopefully this might encourage doctors towards CME (and their administrators towards paying for it). Might a high DIL factor be associated with better treated patients and a higher satisfaction all round?
Last amended: 7:55 AM on 23/02/96