Challenges to alcohol and other drug discussions in the general practice consultation

Abstract

Background. There is a widely held expectation that GPs will routinely use opportunities to provide opportunistic screening and brief intervention for alcohol and other drug (AOD) abuse, a major cause of preventable death and morbidity.

Aim. To explore how opportunities arise for AOD discussion in GP consultations and how that advice is delivered.

Design. Analysis of video-recorded primary care consultations

Setting. New Zealand General Practice.

Methods. Interactional content analysis of AOD consultations between 15 GP’s and 56 patients identified by keyword search from a bank of digital video consultation recordings.

Results. AOD-related words were found in almost one-third (56/171) of the GP consultation transcripts (22 female and 34 male patients). The AOD dialogue varied from brief mention to pertinent advice. Tobacco and alcohol discussion featured more often than misuse of anxiolytics, night sedation, analgesics and caffeine, with only one direct enquiry about other (unspecified) recreational drug use. Discussion was associated with interactional delicacy on the part of both doctor and patient, manifested by verbal and non-verbal discomfort, use of closed statements, understatement, wry humour and sudden topic change.

Conclusions. Mindful prioritization of competing demands, time pressures, topic delicacy and the acuteness of the presenting complaint can impede use of AOD discussion opportunities. Guidelines and tools for routine screening and brief intervention in primary care do not accommodate this reality. Possible responses to enhance AOD conversations within general practice settings are discussed.

Introduction

Alcohol and other drug (AOD) abuse is a major cause of preventable death and morbidity globally.1,2 Many OECD countries including New Zealand3 show a pattern of increasing lifetime alcohol use, decreasing tobacco use and emerging use of other substances of abuse. Primary care is the interface where public health policy initiatives are implemented at a personal health level, and clinical tools and guidelines for GP detection and management of AOD problems have been developed for local use. Identification of AOD risk is important because there is evidence that early detection and intervention is beneficial.4–6

In a New Zealand general practice setting, for example ∼20% of patients will respond positively if asked whether they feel the need to cut down their smoking, nearly 11% will admit to the need to cut down their drinking and 2.8% to cut down on their other drug use.7,8 A medical problem may facilitate AOD discussion by providing legitimacy to introduce the topic,9,10 but social factors also influence how the consultation is constructed.11 As a result, fear of harming the doctor–patient relationship may compound any lack of confidence in addressing some AOD topics.12,13

There is some uncertainty amongst General Practitioners about their role in the delivery of public health initiatives at the personal health level14,15 and this, together with time pressures and the sensitivity of the topic,16 can lead to barriers to AOD discussion in the consultation.12 Recordings have been increasingly used for research into primary care interactions,17–19 and analysis of individual doctor–patient AOD interactions can provide some understanding of how that is played out, successfully or otherwise.9,10,20 For example, a Seattle study used audiotapes to analyse doctor–patient discussions, finding that not all opportunities were taken to explore patient disclosures about alcohol; provider discomfort was evident where there was alcohol-related discussion; and that practitioner advice may be vague or tentative.12 However, published research into how AOD-related issues are actually discussed within the GP consultation remains scarce.

This New Zealand study uses a sample of video-recorded primary care consultations to explore the following questions: How do opportunities for AOD discussion arise and develop in GP consultations? When AOD discussion does occur, what are the structural elements that promote and inhibit those discussions? How do GPs deliver advice for early intervention and harm reduction?

The recorded consultations used for this project are a subset of a larger dataset that was collected for two prior projects using interactional data: a study of clinical decision making in general practice and surgical consultations11^,19 and a longitudinal study where a cohort of patients were tracked and recorded throughout an episode of care.21 In both cases, participants were informed they were being recorded for the general purpose of studying health communication. No specific mention was made of any clinical topic including AOD matters.

Methods

Consultation selection

This study used video-recorded naturally occurring GP consultations in the Applied Research on Communication in Health (ARCH) Corpus at University of Otago, Wellington. The ARCH Corpus is a digitized collection of health interaction videos and related data: transcripts, content summary logs, field notes, participant demographic data and the associated clinical records. Demographic data collected at the time of recording included patient age, sex, ethnicity, educational qualification (by patient self-report), and whether or not the patient was known to the GP. Consultation length was calculated subsequently.

For this analysis, a consultation subset was identified using keywords to search the GP consultations held in the Corpus (171 at the time the study was undertaken), for clinical topics where discussion of AOD was likely to take place.22 The search topics were lifestyle discussions, mental health (depression, anxiety), physical health (liver, gastric or heart problems), injuries and named substances of abuse. Initial search results were scrutinized by a clinician (HJM) to ensure that the final subset would all contain AOD-related interactions.

Consultation analysis

Qualitative methods were used to analyse the doctor–patient interaction in the target consultations. The logs, transcripts and video recordings in the AOD consultation subset were first analysed for content: the location of any actual AOD talk, who initiated it, how long it lasted, whether the patient had admitted to any substance use and which substance(s) they used, whether the clinical discussion was assessment related or management related, if advice or follow-up was offered and, if so, what. Three medical researchers independently analysed content before cross-checking. The consultation data was then analysed in more detail, including multiple viewings of the video recordings, to identify possible interactional facilitators and barriers to AOD discussion. Analysis was first performed by team members independently then findings compared, triangulated and collated in consultation with the wider team. Interpretation of findings was complemented by review of clinical records (where available) and interviews with selected GPs and representatives of primary health care organizations.

Primary care interviews

Three of the 15 GPs in the videos were semi-purposefully selected for interview (by LC), based on doctor gender, practice location and population socio-economics. The GPs had given permission for further contact at the time of consent to the study. In addition, two representatives of primary health care organizations were interviewed (by LC) to establish a primary care perspective on data interpretation. The interview schedule is shown in Table 1.

Table 1.

Interview schedule

1. What are the opportunities to talk about AOD in a GP consultation?

2. How often is AOD talked about in GP consultations?

3. Under what circumstances is AOD usually raised in the consultation?

4. What type of AOD advice is given in the consultation?

5. How valuable do you believe AOD advice is, when given in the GP consultation?

6. How are AOD issues usually screened for in new patients/known patients?

7. Who in the practice usually deals with AOD problems in the patient?

8. What do you see as reasons that AOD opportunities are missed?

9. What do you think are barriers to the discussion of AOD with patients?

10. What are your own feelings towards holding an AOD discussion within the GP consultation?

Results

Topic frequency

AOD topic search terms were found in almost one-third (56/171) of all GP consultations in the ARCH Corpus at the time. Of these 56 consultations, 86% (48/56) included overt mention of drug or alcohol use. Amongst consultations with any mention, 88% (42/48) went beyond a single brief question/comment and paired response, thus some degree of AOD-related dialogue occurred in 75% (42/56) of the subset where AOD discourse would be expected to occur on the basis of the presence of key words, almost 25% (42/171) of all the consultations.

Participant demographics

The subset consisted of 56 individual patients consulting with 15 GPs. The GPs were aged 30–54 (at the time of consultation), 6 female and 9 male, from European, Indian and African continent ethnicities. Their patients, 22 women and 34 men, had an age range of 18–80 years. Self-stated patient education ranged from primary school leavers to those with tertiary education including professional qualifications. The consultation length varied from 04.29 to 35.38 minutes, average length 13.57 minutes. Of the 171 consultations, 34 were first encounters with the GP, 125 were patients known to the GP (12 status unknown), hence 73% of consultations comprised patients known to the GP, with some known to the practice although new to the consulting GP.

Consultation analysis

Alcohol, tobacco, caffeine use and misuse of anxiolytics, night sedation and analgesics were discussed in dialogue that varied from brief questioning (Example 1) to more comprehensive screening, discussion and advice (Example 2). There was a hierarchy of the frequency of topics discussed, with tobacco and alcohol discussion featuring most often, and caffeine, misuse of anxiolytics, night sedation and analgesics less so. All smoking discussions carried an implicit assumption that this referred to tobacco. Some interactions alluded to substance use, such as mention of evenings spent in a public bar, but did not explore the extent of drinking or raise binge drinking as a clinical possibility. Only one consultation, Example 2, included a direct enquiry about (unspecified) drug use. (The original transcripts used a modified form of conversation analysis convention; however, as the analysis in this article focuses mainly on the content of the discussions, the excerpts presented here are in verbatim format, with some punctuation added for readability. Contextual comments are bracketed including some vocalizations, for example tongue-clicking that is transcribed as ‘tut’.)

• Example 1. Brief questioning

• (female patient, age 26; male GP, age 47)

• Consultation code: TS-GP14-02

• GP:    right okay um (tut) do you smoke?

• PT:    no

• GP:    ever smoked?

• PT:    no

• GP:    hah hah and do you drink?

• PT:    once in a blue moon really

• GP:    bl- hah hah

• BOTH:     (laugh)

• PT:    yep never really heavily so

(The consultation then continued on to address another topic.)

• Example 2. Comprehensive AOD discussion

• (male patient, age 19; female GP, age 30)

• Consultation code: TS-GP15-02

• GP:    (tut) alright and um oh we normally just ask a few few questions to see how people are and how things are going

• PT:    yeah

• GP:    hah hah er do you smoke?

• PT:    nope

• GP:    nope that’s great and other things

• PT:    I drink

• GP:    take drugs?

• PT:    n- I drink but nah nah but I don’t do drugs

• GP:    okay that’s good, would you drink a lot or mm?

• PT:    nah not really ‘bout once a week maybe, oh yeah but not never not much though not much

• GP:    when’s it friday night or saturday night

• PT:    yeah maybe those going out nights yeah

• GP:    and do you tend to get drunk on those evenings?

• PT:    sometimes but yeah

• GP:    sometimes, would how m- what do you drink what’s y-

• PT:    oh mainly drink vodkas

• GP:    vodkas

• PT:    yeah

• GP:    how many would you have?

• PT:    twelve bottles

• GP:    (in surprised tone) twelve bottles of vodka?

• PT:    oh nah oh nah

• GP:    shots? hah

• PT:    nah nah nah nah I’m talking about like you know like a like a twelve- er um twelve pack thing like you know bottles li- not like vodkas like you know straight vodkas yeah like lolly drinks yeah yeah

• GP:    oh I see I know what you mean yes, twelve of those?

• PT:    yeah

• GP:    okay

• PT:     hah hah

• GP:    and do you do you drive after that or do you have someone drive you home?

• PT:    nah we maybe catch the bus cos we normally go town sometimes yeah

• GP:    (quietly) okay

• PT:    bus or taxi, nah never drink and drive

• GP:    and do you sometimes not remember what happened the night before?

• PT:    yeah I do but some of the boys don’t they get too wasted

• GP:    yeah

• PT:    mm

• GP:    and have you ever done anything silly because of the drinking?

• PT:    nah

• GP:    (quietly) okay (inhales) and otherwise your mood’s quite good haven’t been feeling low or anything like that?

(Explanatory notes: ‘lolly drinks' are colourful pre-mixed alcoholic drinks often mixed with sweet fruit juice; ‘wasted' is a slang term for being being drunk.)

(The consultation then continued on to address other topics.)

Some examples of this consultation subset included multiple opportunities to discuss AOD topics. In Example 3, during a consultation lasting 25 minutes 47 seconds, a new patient revealed a Hepatitis C diagnosis, then mentioned heavy drinking and smoking, and revealed later his motivation to stop smoking:

• Example 3. Multiple opportunities presented

• (male patient, age 43; male GP, age 39)

• Consultation code: TS-GP03-19

• (at 4 minutes 56 seconds in)

• PT:    um (clears throat) (tut) I still had um hep c I was diagnosed with hep- hepatitis c in two thousand and two

• GP:    mm hmm

• PT:    um (drawls) and since then I’ve had a um well since the non-hodgkins lymphoma I’ve had a um couple of check-ups and I’m okay

• GP:    mm mm

PT:    um (tut) um because of um my depression um which is nowhere near as um significant as what other people suffer from it’s still enough to get me on- have to be on medication all the time um I’ll probably be on it for the rest of my life, I think um I I do suffer from major bouts of paranoia and you know it’s chronic you know it’s really bad

• GP:    what kind of paranoia what kind of things do you worry about?

Paranoid thoughts were subsequently explored but the underlying reason for HCV infection was not discussed further on this occasion (although the doctor did mention twice elsewhere in the consultation, that this was an issue to be addressed in the future). The risks of drinking with hepatitis were not raised when patient admitted to smoking and drinking to cope with the strain of a relationship breakdown:

• Example 3 (continued)

• (at 7 minutes 29 seconds in)

• PT:    not that I remember um I just what happens (drawls) is I um I buy strong cigarettes um cigars, smoke them like cigarettes and I buy buy alcohol

• GP:    mm

• PT:    wine or beer and I go through about five cigars with a couple of bottles of beer or about five glasses of wine to um in the evenings to get me on this on a level (tut) and then and then that gets me back to a s- um I spend a reasonable amount of time by myself in the evenings

• GP:    mm

• PT:    you know like I’ve just been I’m just coming out of a r- what I call a relationship um well it is a relationship I’ve I was overseas and my (inhales) (tut) um my partner was back here in New Zealand working

• GP:    mm

• PT:    and we’ve just um we’ve just split up um mainly because I think I can’t handle um well the relationship puts a lot of strain on me

• GP:    mm

The alcohol confession was accepted without further exploration, and no specific AOD advice was given. Other health complaints were mentioned including chest pains (at 11 minutes), but the patient suggestion that this may be linked to smoking was not explored:

• Example 3 (continued)

• (at 10 minutes 5 seconds in)

• PT:    we can do this and um I value my work and I also value myself and and that I want to do the best you know and um, I’m already, I’ve got slight chest pains as a result of my smoking and um yeah it’s it is through my smoking and on and off cigare- um strong cigarettes like tobacco that I was smoking when when I was overseas is that french cigarettes and they’re triple the um the

• GP:    tar and

• PT:    the tar and everything that you get in new zealand you can’t buy that in new zealand any more so I don’t do things by half measure you know um at all and in this relationship I do- either like to be in it or out of it and I

• GP:    mm mm

• PT:    I wanna be her friend and it’s not really easy for her at all it sucks for her actually she’s someone that I’d like to spend the rest of my life with but at the moment I I just feel (tut) that I’ve gotta stand on my own two feet you see that’s that’s gonna be re-another way

• GP:    have you er done much counselling work yourself around some of these issues?

• PT:    I haven’t um I did see a counsellor bef- this year but I I just felt like I was um um a cookie in a jar really I just felt like I was yeah I I’m maybe I’m too sensitive but um um I I do pick up on you know I do go with I do pick up on how people speak to me and (exhales) how they listen and um body language and everything just like anyone else

Later on in the same consultation, this patient further discussed his anxiety, panic attacks, skin problems and longstanding mental health issues including depression and completed a depression score sheet. The patient had presented a big agenda for a first-time consultation. The focus of this consultation on this occasion was clearly relationship building, where the GP appeared to have prioritized active listening and relationship building, at the expense of in-depth history taking, advice, brief intervention or even reinforcement of the stated motivation to stop smoking.

Requests for prescription medication with abuse potential such as sedatives presented opportunities for AOD discussion. An impending initial admission (in Example 4) about anxiolytic use, ‘I’m us-‘, was re-formulated. The GP supplied reasons why the patient might wish to continue with benzodiazepines at an increased dose ‘… you’re still not feeling great?’ ‘You get a bit anxious and stressed …?’ The self-administration regime and total daily dose remained unstated, but the GP attempted to negotiate a smaller dose increase than requested. This drew an ambiguous response, hinting at unspecified changes that the patient was going to ‘have a few goes at’:

• Example 4. Dose increase requested

• (female patient, age ∼66; male GP, age 46)

• Consultation code: TS-GP09-20a

• PT:    (inhales) oh I’m us- you know those tablets I used to get my nerves and all that? Those ones there, can I use two instead of one, or no?

• GP:    you (drawls) probably- you’re still not feeling that great? you get a bit anxious and stressed at times?

• PT:    um yeah I, I don’t sleep that well but

• GP:    I think you could go up to one and a half and see how that goes?

• PT:    (no)

• GP:    you wouldn’t wanna jump straight from one to two might feel a bit weird but you could go up to one and a half it would mean having to cut them in half

• PT:    yeah

• GP:    you could try that for a couple of weeks

• PT:    yeah (mumbles) oh no I’ll try and leave it like that see how it- have a go- a few goes at it

The GPs in examples 4 and 5 both offered advice about appropriate medication use, and in Example 5, the GP also used the opportunity to warn of addiction risk before completing the prescription.

• Example 5. Sleeping pill request

• (female patient, age 25; female GP, age 36)

• Consultation code: TS-GP08-22

• GP:    I’m very happy for you to have a few zoplicone tablets (inhales) um to take um (types) I guess just use them (exhales) you know er- sensibly because it is a potential

• PT:    mm

• GP:    to get addicted these if you were to use them every night so (inhales) you know certainly use them every night for the next few nights

• PT:    mm

• GP:    but after that you know just take them when you you really need them and you may find it’s just around the times

• PT:    mm

• GP:    that you’ve been doing night shifts that they’re most useful

Open-ended questions for initiating AOD enquiry were observed to be very successful, as in Example 2 where the GP first asked in an open-ended manner how things were going and then enquired more specifically about smoking, drug use and drinking. Example 2 also contained closed questioning about mood (on the last line of the excerpt). Closed statement AOD questioning was generally negatively polarized: ‘you haven't been a cigarette smoker?’ or ‘you’re not drinking much?’ Example 6 shows successive negatively polarized closed statement questions about both personal smoking and smoking associates, leaving the actual patterns of use unspecified. This communication tactic permitted ambiguity on the part of the patient, and it may also facilitate denial of the extent of substance use, although in this case the tone of the consultation: ‘clean little lot you two’ suggests pre-existing familiarity and understanding:

• Example 6. Closed statement questioning and ambiguity

• (female patient, age 68; male GP, age 45)

• Consultation code: TS-GP09-02

• GP:    and you’re not you don’t associate with anyone else who smokes, you’re not around anyone that smokes, so you don’t really (inhales) just you and (NAME of patient’s husband) really you don’t have too much?

• PT:    he doesn’t have smokes now either

• GP:    no no clean little lot you two

Many of the consultations exhibited a degree of interactional discomfort or delicacy during the AOD talk sequences, with verbal and non-verbal body language discomfort of both the doctors and patients captured on the video footage. Although pausing, over-talking, rephrasing and incomplete sentences are normal in spoken interactions, cases of marked verbal dysfluency can be seen as a possible sign of interactional discomfort for the doctor or patient. Verbal cues of discomfort for patients also included strategies such as understating or becoming defensive. Commonly, the GPs did not explore the patient’s understatement or challenge their defence or re-assert any health promotion message. This reaction could be designed for relationship building: acceptance of the patient confession at face value, deference to patient sensitivities or postponement of difficult topics for another day. In Example 7, the GP did not challenge the confession of an occasional smoking ex-smoker nor enquire further about alcohol intake in the pub or elsewhere:

• Example 7. Patient understatement

• (male patient, age 31; male GP, age 45)

• Consultation code: TS-GP10-07

• PT:    oh I did smoke

• GP:    yeah

• PT:    I gave up for about five years

• GP:    right

• PT:    um I do have a cigarette every now and again when I’m in the pub

• GP:    okay

• PT:    but I’ve kind of done that for ages it’s not a new thing

• GP:    so it’s not a new thing yeah

• PT:    yeah I’ve I don’t I w- yeah I don’t buy them I wouldn’t say I was a smoker

Some consultations showed the doctor providing support and brief advice in the face of heavy drinking, to keep the defensive patient on-side and thus potentially avoiding confrontation. GPs sometimes used positive interjections to give the patient permission to continue, as in Example 3, and Example 8 shows how this facilitated the ‘confession’:

• Example 8. Patient defensiveness

• (male patient, age 52; male GP, age 52)

• Consultation code: IS-GP01-02

• GP:    I do think you know if you could look at certainly cutting down a fair bit um (exhales) and you know I mean I would have thought if you if you’d managed to get down to half of what you’re drinking now on a regular basis then there’s that’s still you know sort of half a bottle of gin um it’s not as if you’re going without

• PT:    well it’s probably I’m exaggerating

• GP:    okay

• PT:    that’s a big night I might have six seven

• GP:    okay

• PT:    corona and lime

• GP:    yep

• PT:    which isn’t

• GP:    all right

• PT:    and then I might have half a dozen doubles of gin or something

• GP:    okay so (pause) you you reckon it was sort of ten minutes or so a couple of days

• PT:    probably it mightn’t have been that long, for this to go?

Topic changes were another tactic employed by GPs to defuse sensitivity and acknowledge interactional delicacy. In Example 8, the GP ends the alcohol discussion by returning to the presenting complaint without any further comment on the patient’s summary of his alcohol intake. Some patients appeared complicit with that tactic and (as seen in examples 2, 5 and 11) did not challenge that. However, in Example 8, the sudden topic switch appeared to take the patient by surprise, prompting him to point to his presenting sore eye. Even when the AOD topic was first raised by patients, some GPs understated the importance of the problem, recommending intake reduction and rolling with resistance, as in Example 9.

• Example 9. Doctor’s understatement

• (male patient, age 79; male GP, age 31)

• Consultation code: TS-GP10-05

• GP:    I think your liver’s probably complaining a wee bit maybe from the alcohol (inhales) um yeah

• PT:    oh I’ll grizzle any anyway if it’s er er if I er (drawls) I I must have been a potentially alcoholic (inhales) you know I grizzle if I don’t have it and I grizzle if I do

• GP:    if you do (laughs) yeah so I think we need to check that out to

• PT:    yes

• GP:    make sure there’s no damage going on there

• PT:    yes yes

• GP:    and um really just go from there depending on what we find but if you can try and pull that er the old alcohol back a wee bit maybe you know one or two a night

• PT:    oh yeah I think (mumbles) they w- th- the bottle of whisky (in a high pitched voice) where’d I get that anyway? somebody gave it

• GP:    (laughs) somebody gave it to you

Shared humour can be face-saving, and this was used to facilitate confessions of excessive drinking as seen in Example 9 above and Example 10:

• Example 10. Shared humour

• (female patient, age 27; male GP, age 48)

• Consultation code: IS-GP02-07

• PT:     but I guess disinfecting yourself with vodka is not really helpful is it?

• GP:     (tut) it’s tempting

• BOTH:     (laugh)

• GP:     very tempting

• PT:     (laughs)

• GP:     but um no probably in the long run it’s not going to er not going to help much no, a little bit of alcohol maybe but er not a lot

• PT:     not half a litre

• GP:     no no no half a litre is probably overdoing it

• PT:     (laughs) okay

However, in Example 11, the shared humour ceased in the face of persistent serious questioning from the GP:

• Example 11. Navigating humour

• (male patient, age 66; male GP, age 47)

• Consultation code: TS-GP14-04

• GP:    and have you (pause) thought of quitting the smoking?

• PT:     hah (laughs) oh I think about it all the time yeah

• GP:     hah hah hah hah hah hah hah but what action are you taking?

• PT:     oh I just give up

• GP:     just thinking?

• PT:     I give up but you know then I’ll feel like a cigarette and I’ll run round and get another bloody packet, silly

• GP:     do you want to talk with the quitline or?

• PT:      no if I can give up I’ll just give up

• GP:     oh

• PT:      oh I don’t worry about that, just a bit of will power that’s all, I can’t think what I was down here for (sighs), isn’t it silly? Doctor (NAME of another GP) put me on a course of six pills for …

(The patient then diverted the topic away and neither GP nor patient subsequently revisited it.)

AOD advice was sometimes given in a mitigated and non-specific manner, for example to ‘pull back a wee bit’ (Example 9) or ‘keep under control’ (Example 12):

• Example 12. Mitigated advice

• (male patient, age 40; male GP, age 31)

• Consultation code: TS-GP10-03

• GP:    at the moment, just get a bit of a baseline from that point of view um and ye- if you’re a non smoker um not drinking a hell of a lot weight’s little bit high

• PT:    yeah

• GP:    but not not massive um so those are the the three major things really to keep under control and you’re obviously ah keeping a a good eye on that at the moment

• PT:     mm

• GP:    so we’ll get those checked and go from there

The most complete AOD discussions were quite prolonged interactions (as in Example 2), but long consultations did not guarantee that any AOD discussion would take place, as we saw in Example 3. Some shorter discussions demonstrated motivational interviewing techniques of rolling with resistance, handling the interactional delicacy and relationship building.

Primary care interviews

All three GPs who were approached for the interview to explore these findings granted the request. Each identified time pressure and the sensitivity of AOD topics as the main barriers to effective AOD discussion. A primary health organization manager also explained that the user-pays primary care system offers no incentive for doctors to spend additional consultation time discussing AOD problems that might arise: an unmotivated patient is unlikely to want to return at a later date and pay more to discuss AOD issues further, and not all shared care schemes for primary care review of suitable mental health patients extend funding to include substance use problems.

Discussion

These findings are consistent with other studies that have identified that doctors do not always take up opportunities to discuss AOD with patients9^,12^,23^–25 even when the patients themselves voluntarily divulge information about that.12,24 However, the prevalence of AOD dialogue in this study was surprisingly high given the context of naturally occurring general practice consultations. Many possible factors impact on whether or not AOD discussion opportunities are taken up by GPs and their patients.

In the subset of consultations analysed for this paper, AOD use itself was not the primary presenting complaint, although it could be inferred to have been a factor in the presentation in some examples. AOD topics were generally introduced in the context of exploring presenting symptoms, systems enquiry or general health screening. Smoking discussions occurred more often than alcohol, but in general, smoking dialogue also appeared to be more comfortable for both parties than alcohol dialogue. Factors contributing to the uptake of opportunity for an AOD discussion can be grouped into three broad categories: Interactional, Clinical and System or Policy factors.

Interactional factors include the interpersonal skills of both GP and patient, verbal and non-verbal communication and the perceived delicacy of the topic. The GPs themselves mentioned the sensitivity of the topic as a potential barrier to discussion when interviewed. In this study, the GPs typically enquired about AOD use in non-threatening ways, used a mix of open and closed statement questioning styles, and sometimes put forward statements for patient agreement. Patients typically minimized, rationalized or gave defensive or socially acceptable answers. GPs accepted patient statements largely without challenge, although the analysis suggests that this stance can be justified, for example to build relationships. Video footage also captured body language reflecting discomfort. The sociological concept of ‘face work’26^,27 provides a useful explanatory framework for this finding.

In this study, cases of apparent avoidance of use of opportunities to discuss AOD were often accompanied by complex interactions of ’face saving’ on the part of both doctor and patient, a feature of consultation behaviour that is not limited to AOD discussions, nor to GP consultations.11^,26^–30 Such interpersonal face work may protect the integrity of the patient and may also support the GP where confidence in AOD knowledge is not strong16 or when the nature of the personal interactions make the GP reticent to enter the discussion on this occasion.19

Although both GPs and patients can be observed to orient to the importance of AOD use and its impacts on health, there are often challenges inherent in the discussion. This may be signalled by apparent dysfluency in the consultations, which is in part a normal feature of any spoken interaction, but in excess can be indicative of an interactional dilemma.19 Similarly, use of humour may reflect skilful targeting of the level of interaction to appeal to the patient’s degree of insight and motivation. Socio-legal constraints may add to the observed interactional delicacy since patients may naturally be reluctant to divulge and GPs reluctant to explore, illicit activity or socially unacceptable behaviour. Prior studies also note that smoking and drinking advice may be given and received differently12^,23 and the discomfort of GPs in handling an AOD discussion with patients has been considered as a contributory factor.12,31

Clinical factors can also hamper AOD discussion. General practice is a wide ranging specialty branch of medicine demanding great breadth of knowledge and skills. Consultations are also complex and multi-faceted—during any consultation GPs actively triage multiple issues. The interviewed GPs mentioned that time pressure and the obligation to primarily manage the presenting complaint determine that AOD topics cannot always be attended to immediately, even when raised. Literature about competing demands in the general practice consultation32 has identified that the potential of the GP to engage with preventive activity is unlikely to be activated on all occasions.33 This phenomenon could alternatively be regarded as ‘mindful prioritising’34,35 where refraining from always following up on AOD issues represents a sensitive and patient-centred approach to develop and maintain the doctor–patient relationship. While smoking and alcohol consumption are more common behaviours than other recreational drug use in New Zealand and other OECD countries,7,36 the lack of discussion about other drug use is an important issue for further debate.

Opportunities for AOD discussion may also be difficult to take up where a third party such as a carer/relative/partner/friend/support person also attends the consultation, since the GP may have to judge if an AOD discussion is appropriate in the presence of that third party. Third person support is often encouraged in health consultations, but this consultation subset did also include examples where the third party (a spouse or caregiver) had introduced additional issues of their own, impacting on the GP's attempt to engage the presenting patient in an AOD discussion. The role of third persons as a potential distraction from engaging in AOD discussion deserves further investigation.

System or policy factors include workplace practices and policies, systems for managing routine screening and consultation time, funding, referred service availability, eligibility criteria and waiting lists. In New Zealand, the primary care system retains a large ‘user-pays’ component, making funded follow-up particularly difficult in circumstances such as AOD where the patient might require some motivation.

Limitations

The methodology applied here cannot open up the ‘black boxes’ of clinical reasoning and the patient’s intended agenda. It was not possible to ascertain, from analysis of the data, why AOD topics were not always raised or pursued and why advice was not always given or early intervention offered. Some of the apparent missed opportunities may have occurred for very good clinical reasons.

The sample involved a limited number of consulting GPs and their patients and an even smaller number of GPs were interviewed regarding the interpretation of findings. Therefore, the extent of generalizability and transferability is unclear. The data examined for this study provides only a snapshot, a single consultation in each case, and therefore lacks information about AOD enquiry or intervention that may have taken place subsequently or previously. Although no mention was made of AOD at the time of data collection (as this was not the focus of the studies from which this data was drawn), video recording of consultations may have introduced bias to this particular study if it discouraged patients wishing to discuss illicit drug use topics with their GPs from participation or encouraged them to opt to not discuss drug use on that recorded occasion. Patients who decide to reveal illicit drug use to their doctors need to trust in the privacy and confidentiality of their medical information.

This study does not establish whether the apparent compromise between a patient-centred approach and the need to routinely screen for AOD use in some consultations is appropriate or inappropriate, nor can it judge appropriateness of the advice given. It is difficult to judge to what extent these findings represent active avoidance or mindful prioritization under the pressure of competing demands, but the interviewed GP’s indicated that both phenomena may be occurring.

Conclusions

This study has identified high levels of interactional skill in AOD consultations and also structural challenges which may lead to apparent ‘missed opportunities’ for AOD intervention and AOD-related health promotion. The findings affirm the concept of ‘mindful prioritizing’ as a valid description of consultation dynamics alongside that of ‘competing demands’.

Within this small sample of consultations, there were instances where the clinical interaction and advice given did not follow existing New Zealand primary care guidelines.36^–38 However, this finding is not unique to GPs or to AOD discussion. Non-use of guidelines and protocols is seen elsewhere in clinical practice21 and there may well be good reasons for seemingly ‘bad practice’.34,39 Possible reasons for this should be explored to inform future strategies to enhance AOD conversations within general practice settings. It may be necessary to explicitly explain to trainees the patient-centred reasons for apparent omissions from the consultation and to ensure that practices have systems in place to ensure that postponed opportunities for AOD discussion will be followed up at a later appointment. Guidelines and tools for screening and brief intervention in primary care carry the expectation that GPs will use all opportunities to discuss AOD in primary care, but these naturally occurring consultations have shown that this expectation is unrealistic. Re-design of clinical guidelines to better fit the patient-centred approach used by GPs to develop and maintain therapeutic relationships over time with their patients may be beneficial.

The findings also carry implications for vocational trainers and providers of continuing professional development programmes to incorporate resources to help to build the clinical skills needed to facilitate use of opportunities for AOD discussion within the consultation.

Declaration

Ethical approval: The subset of consultations used for this project were collected for two prior projects: the Interaction Study which was approved by the Wellington Ethics Committee reference no. 03/09/090 for the project entitled ‘Exploring clinical decision making when rationing is explicit’, and the Tracking Study ‘Tracking healthcare interactions: patient-professional communication’, approved by Central Region Ethics Committee reference no. CEN/05/12/096. Approval for analysis of the subset for the purpose of this current project was obtained from the University of Otago Ethics B process for studies involving human participants.

Funding: New Zealand Health Research Council; the Royal Society of New Zealand Marsden Fund; the New Zealand Ministry of Health National Drugs Project Development Fund.

Conflict of Interest: None to declare