Table 2 Features of an acceptable alcohol screening tool for general practice, clinicians’ perspectives
Suggestion | Feature description or justification | Participants’ comments |
|---|---|---|
Automated documentation of risk stratified scores | - Helpful to generate scores to risk stratify patients - Risk scores need to be automatically and appropriately documented o So other clinicians can find them o Because if it’s not automatic, it probably won’t be adequately recorded | “We're all prompted to put a record into My Health Record. Do we all do it? No, because we don't have time. We try our best… but I don't think we hit the targets for doing the uploads. It would make far more sense to have it (alcohol risk scores) automatically uploading… practising defensible medicine and being able to document everything is vitally important. Have a prompt that is going to do a lot of the work of recording as well.” (GP03) |
No overwriting of scores | - Allow assessment of a patient's alcohol risk profile across different timepoints to see change over time for reflecting on management progress | “The problem with smoking assessment too (as well as recording of alcohol consumption), is that it doesn't record previous consumption or use. You can only put in what they do now and when you update it the previous is gone and … if they used to be a very heavy drinker, but now they drink very rarely or not at all. That is not captured very well in the data.” (GP College rep) |
Field auto-population | - Auto-populate relevant fields in health summaries, electronic referrals and shared maternity care records (with patient consent) and thereby streamline workflow to avoid duplication of effort | “So if it auto-populates when we need to do a referral that is very helpful because it just saves you time.” (GP02) “I think what’s really helpful, if it could automatically link into the referral, for instance. Click on the link and up comes your template for the referral, that would be great.” (GP03) |
Data accessible for reporting | - Having reports of the captured data readily accessible to support quality improvement activities (noting that quality improvement activities are necessary to maintain accreditation, and some practices also required this to maintain funding) | “Because of our CPD that we have to do “Measuring Outcomes”; isn’t that going to be so much easier if you can do a nice search for your scores? An audit for someone could be: In the last thirty antenatal cases that you saw, how many did you do drug and alcohol screening? And it's automatically there, with your tool. (GP03) |
Provide advice that is high level and non-binding | - Advice provided by the tool should not include specific recommendations that may not be feasibly actioned, or referral pathways that are not locally available - Management for individual patients is best decided by clinicians who know their patients and the availability of local services, not by the tool | “We have one psychologist, and for adults it's over 12 months wait. And a psychiatrist its 18 months, so having a pop up that says “Refer to psychiatrist” is actually not helpful at all. What you do with it I guess is still up to the clinician.” (GP04) “You’re not going to be able to make resources for every single clinical context because you don’t know who is sitting in the room with the doctor and you don’t know what the resources are that the doctor and the patient have at their disposal.” (GP10) |
Include only a few key resources | - Include one or two contacts for peak organisations - Unfeasible to maintain local referral resources given the constantly changing landscape of services with funding cycles | “It would be very hard I guess for you guys to roll out a thing that's localised. Like for me, if I need mental health support, I'd be calling our local mental health triage number, but that's very different to what you would do if you were in (another city), for example.” (GP04) |
Suitable for a range of communities | - Ensure the tool is optimised for a range of communities, including regional and remote, not just metropolitan areas of the “southern states” | “If you were talking about rolling the software out to places that have (Aboriginal) communities, then I would do a big consultation process. Get a council of Elders together to ask them whether they think it's appropriate to ask these questions. And of course, what you might ask in Sydney is going to be different to the Northern Territory.” (GP06) |
Accessible outside of automated prompting | - There should be more than one way to access the tool. The tool should be readily accessible on the computer desktop or easily located in the clinical software, for opportunistic screening such as when a patient is planning pregnancy | “That should be a regular thing that we should be screening, even before getting pregnant, of course. But if not, then reassessing again when someone is pregnant.” (GP02) |
Passive prompts | - Avoid ‘alert fatigue’ due to forced interaction with prompt. Prompts should not be interruptive of workflow | “Some people don't even look at the prompts and just put, you know, cancel, cancel, cancel.” (GP02) “It would be a barrier if you couldn’t move onto the next page (of the tool) without filling it in.” (GP) |
Repeated prompts | - Repeat prompts, for example, towards the middle and end of a pregnancy, could be useful to guard against assumption that alcohol use status pre-pregnancy or in early pregnancy has not changed - When care is shared between general practice and antenatal clinics, prompts could be scheduled to fit the visit schedule | “I think it's a good reminder, when you put in your antenatal check, but it would be really good to have the option of using it… prenatally or even in the middle of a pregnancy or postnatally as well.” (GP01) “If we are doing shared care, then we have set times when we see them with their local antenatal clinic. So that's when we would do them (screening).” (GP02) |
Multifunctional rather than single purpose | - A large volume of risk relevant information needs to be gathered in an early antenatal consult, such that some questioned the sense of integrating a tool just for alcohol screening - It was suggested that an antenatal clinical decision support tool could, as a minimum, include screening for alcohol, nicotine and other substance use, but might also include elements of a psychosocial screen, including mental health, domestic and family violence, nutrition, housing - Framing discussion about alcohol use within a broader assessment framework could be appealing to some clinicians | “The decision support on alcohol intake is but one part of a whole bigger issue that needs to be dealt with.” (GP College rep) “If we add nutrition into that as well, you ask are they taking folic acid, tick the box, you've asked about diet, tick the box, you asked are they working, tick the box. Having everything built into that would be great.” (GP03) “It doesn't make sense to me… to have a tool that only asks about alcohol … It's taking one thing when screening for domestic violence is important, smoking is important, not eating soft cheese that might give your baby Listeria is important. So it doesn't work from a practical point of view, having alcohol by itself. (GP academic) “You need to include other drugs and when we're talking psychosocial, to me, it's housing, it's food, it's violence, it's alcohol, it's smoking… I would include a bit more of the bigger picture.” (GP06) |
Multifunction with Save and Return feature | - A multifaceted screening tool should have the functionality to partially complete, save and return, to allow clinicians to pivot the focus of their consultations in response to what a patient discloses during screening | “I wonder if rather than the tool being a whole, like you administer everything all at once, having the functionality… so you could do it over like a couple of consults. If it's just too much to do in one.” (GP01) |
If standalone: AUDIT-C for pregnancy | - If the tool was alcohol only, then AUDIT-C for pregnancy was considered appropriate to enquire about and quantify alcohol use. The brevity of AUDIT-C for pregnancy was a strong point | “If it's only a few buttons (clicks on the screen), like the Audit C tool… So long as it's not 10 questions, then I don't think that's a problem” (PN1) |