Providing routine chronic disease preventive care in community substance use services: a pilot study of a multistrategic clinical practice change intervention.

OBJECTIVES: To evaluate the potential effectiveness of a practice change intervention in increasing preventive care provision in community-based substance use treatment services. In addition, client and clinician acceptability of care were examined.
DESIGN: A pre-post trial conducted from May 2012 to May 2014.
SETTING: Public community-based substance use treatment services (n=15) in one health district in New South Wales (NSW), Australia. PARTICIPANTS: Surveys were completed by 226 clients and 54 clinicians at baseline and 189 clients and 46 clinicians at follow-up.
INTERVENTIONS: A 12-month multistrategic clinician practice change intervention that aimed to increase the provision of preventive care for smoking, insufficient fruit and/or vegetable consumption and insufficient physical activity. PRIMARY AND SECONDARY OUTCOME MEASURES: Client and clinician reported provision of assessment, brief advice and referral for three modifiable health risk behaviours: smoking, insufficient fruit and/or vegetable consumption and insufficient physical activity. Clinician-reported optimal care was defined as providing care to 80% of clients or more. Client acceptability and clinician attitudes towards preventive care were assessed at follow-up.
RESULTS: Increases in client reported care were observed for insufficient fruit and/or vegetable consumption including: assessment (24% vs 54%, p<0.001), brief advice (26% vs 46%, p<0.001), and clinicians speaking about (10% vs 31%, p<0.001) and arranging a referral (1% vs 8%, p=0.006) to telephone helplines. Clinician reported optimal care delivery increased for: assessment of insufficient fruit and/or vegetable consumption (22% vs 63%, p<0.001) and speaking about telephone helplines for each of the three health risk behaviours. Overall, clients and clinicians held favourable views regarding preventive care.
CONCLUSION: This study reported increases in preventive care for insufficient fruit and/or vegetable consumption; however, minimal increases were observed for smoking or insufficient physical activity. Further investigation of the barriers to preventive care delivery in community substance use settings is needed. TRIALREGISTRATION NUMBER: ACTRN12614000469617. © Author(s) (or their employer(s)) 2018. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.

The first study to evaluate the effectiveness of a multistrategic clinical practice intervention on the provision of preventive care to community-based substance use clients for multiple health risk behaviours.

A strength of this study is the use of multiple data collection points with both client and clinician report.

The study is limited by the pre-post design; however, this design was considered appropriate for a preliminary examination of the potential effectiveness of a clinician practice change intervention.

The use of participant self-report of care provision may result in overestimates of care provision.

Background

People with substance use problems experience a life expectancy of up to 25 years less than the general population.1–3 A proportion of this reduced life expectancy is attributed to high rates of preventable chronic illness such as cardiovascular disease, diabetes and cancer.1 4–7 Contributing to such illness is a high prevalence of modifiable health risk behaviours8 9 such as smoking, insufficient fruit and/or vegetable consumption and insufficient physical activity.10–14

Clinical practice guidelines recommend that preventive care should be delivered routinely by health service providers to reduce the prevalence of modifiable health risk behaviours.15–21 Although preventive care practice guidelines are not specific to substance use treatment settings, there is increasing research and service delivery interest in enhancing the delivery of such care in this setting.22 23 Substance use treatment services represent an opportunity for the provision of such care as they often involve multidisciplinary care teams who are skilled in behaviour change and incorporate multiple treatment episodes.23–26 Despite this, the provision of preventive care in substance use clinical settings occurs at suboptimal levels.27–30 For example, a study within 111 US community substance treatment programs found that 66% of counsellors routinely asked patients if they smoked, 44% advised smokers to quit and 30% provided patients with a quit-line number.30 The only study identified that reported the prevalence of care for insufficient nutrition or physical activity found suboptimal levels of client reported assessment (22% and 51%), brief advice (25% and 49%) and helpline referral (1% and 2%), respectively, across 15 Australian substance use treatment services.29

Only one study could be identified that examined the effectiveness of an intervention to increase preventive care delivery in substance use settings. Using a pre-post study design, Guydish et al 31 examined the effect of a 6-month practice change intervention on the provision of smoking cessation care in three US residential substance use services. The practice change strategies included consultation, leadership and support, policy development and staff training. Clients (n=150) reported an increase in the receipt of education, pamphlets, pharmacotherapy, counselling and the provision of support groups. An increase in favourable attitudes towards smoking cessation care was reported by both clients and staff (n=143). Although such results suggest that practice change interventions may increase preventive care delivery, the study was limited to a single health risk behaviour and residential settings.

Given the limited extent of intervention research, a study was undertaken to assess the potential effectiveness and acceptability of a multistrategic, evidence-based, practice change intervention to increase the provision of preventive care for smoking, insufficient fruit and/or vegetable consumption and insufficient physical activity in community substance use treatment settings.

Methods

Study design and setting

A pre-post trial was conducted within public community-based substance use treatment services in one health district in New South Wales (NSW), Australia. Clients are referred (including self-referral) to the free government-funded services.

Participants and recruitment

Community-based substance use treatment services

Community-based substance use treatment services were chosen as approximately 90% of Australian substance use treatment occurs within outpatient or community services, with the remaining occurring in residential or inpatient settings.32 33 In Australia, publicly funded outpatient and community services are conducted outside of inpatient or residential settings. In the context of this study, outpatient settings are generally provided by hospitals, whereas community health services are under the remit of community health. Services were ineligible to participate if they were: residential, inpatient hospital or intake-only services; primarily cared for clients under 18 years; or only provided care in a group setting. All eligible public community-based services (n=15) in the study region were invited to participate. The services included those providing counselling and management, outpatient withdrawal, stimulant and cannabis specific treatment, and court diversion programs.

Clients

Clients were eligible for data collection if they: were ≥18 years of age; had a face-to-face appointment within the previous 2 weeks; had not previously completed the survey; and had not been identified as inappropriate for contact by their clinician (eg, were seeing a confidential service and family members or others may become aware of their treatment through the client participating in the data collection procedures). Additional eligibility criteria included English proficiency and mental and physical capability of completing the study procedures (determined by interviewers on client contact).

Each week throughout the baseline and follow-up data collection periods, 30 eligible clients were randomly selected from the substance use services’ electronic medical record system using SAS V.9.3 software.34 Selected clients were mailed a study information letter.

Clinicians

All clinicians received the intervention. Clinicians were eligible to participate in data collection if they: had ≥10 face-to-face appointments with adult clients within the previous 2 months; had been employed by the service for at least 3 months; and were not on leave. Eligible clinicians were mailed an information letter.

Intervention

Model of preventive care

The preventive care model was based on the 2As and an R model (ask, advise and refer).26 This model has been recommended as an effective way to address modifiable health risk behaviours while also acknowledging barriers to care provision, including competing clinical priorities and brevity of clinician contact.19 26 35 Clinicians were required to assess all clients every 6 months (as indicated by the clients’ medical records) regarding their smoking, fruit and vegetable consumption and physical activity. Assessment was recommended to occur every 6 months to emphasise the importance of routine provision of preventive care. Assessment questions aligned with, and client risk status was determined according to, the Australian National Guidelines.36–39 For clients who were identified as ‘at risk’, clinicians were required to provide brief advice and referral to further support. Referral options were free government telephone coaching services: the NSW Quitline and the NSW Get Healthy Information and Coaching Service; general practitioners (GPs) or Aboriginal Medical Services (AMS); and other local care delivery providers (eg, dietitians and physiotherapists). Telephone helplines have been shown to be effective in the general population40 41 and for people with substance use problems.35 In addition, such services provide free and readily accessible support, which may address barriers to treatment for people with substance use problems, such as transportation and cost.42

Clinical practice change intervention

The following evidence-based43–46 practice change strategies were implemented simultaneously in all participating services over 12 months:

Leadership and consensus46: the preventive care model was formalised through a health service-wide policy. Clinical teams and management were consulted prior to and during the implementation of the intervention, at least once a fortnight. Senior health district management were consulted to confirm their continued support for the intervention and to reach agreement on key performance indicators.

Enabling systems and procedures: the existing electronic medical record system was modified to include a health risk behaviour assessment tool to support the standardised provision and recording of care. The tool recorded the assessment of client health risk behaviours (eg, Have you ever been a smoker?) and the client responses. The tool provided suggested wording regarding the provision of brief advice in line with Australian national guidelines36–39 (eg, The best thing you can do for your health is to stop smoking) and suggested referral services in an effort to standardise and prompt care delivery.

Clinician and manager training43: clinicians and service managers were invited to face-to-face training conducted by a support officer for approximately 2 hours and were required to complete six competency-based online modules (approximately 1.5 hours). The training and online modules covered the recommended model of preventive care, providing and recording care, and cultural appropriateness. All eligible clinicians and service managers had attended face-to-face and/or online training at the completion of the intervention.

Implementation support: a support officer with health promotion and clinical practice change experience was allocated to each site, providing the initial face-to-face training, a minimum of one visit per month and fortnightly phone or email contact with managers. The support officer provided training and discussed project progress, any problems or concerns and potential solutions to overcome such concerns. Managers were encouraged to proactively contact the support officers with concerns or problems.

Monitoring and feedback44: each month, service-specific feedback reports regarding the prevalence of preventive care were generated from the electronic medical records and emailed to managers. Support officers discussed the service’s performance and plans for increasing care provision with managers. Managers were encouraged to share the feedback reports with their staff.

Provision of clinical practice change resources45: clinicians and managers were provided with resources including: a guide to providing and recording preventive care in the medical records system; copies of a paper assessment tool to be used when a computer was not accessible; flip charts; client handouts; helpline fax referral forms; posters; and monthly newsletters. An email helpline and internet information site that included printable versions of all resources were also provided.

Data collection procedures

The primary method of data collection was client and clinician computer-assisted telephone interview (CATI) administered by trained interviewers.

Clients

Client surveys were undertaken prior to intervention commencement (baseline: May–October 2012) and immediately following intervention completion (follow-up: November 2013–May 2014). Clients were contacted 2 weeks after receiving the information letter to complete the survey (approximately 25 min).

Clinicians

Clinician surveys were undertaken prior to intervention commencement (baseline: October–November 2012) and immediately following the intervention completion (follow-up: November 2013–February 2014). Clinicians were contacted 4 weeks after receiving the information letter to complete the survey (approximately 30 min).

Patient and public involvement

The focus of the study was increasing the provision of preventive care by clinicians. Clinicians were involved in the development of the study. Patients were not included in the development of the research question, outcome measures, data collection procedures, recruitment or conduct of the study. The results will not be disseminated to patients.

Measures

Participant characteristics and health risk behaviours

Client age, gender, postcode and number of appointments with the service within the previous 12 months were obtained from electronic medical records. During the CATI, clients were asked their: employment status, Aboriginal and/or Torres Strait Islander status, marital status, highest level of education attained and any mental or physical conditions for which they were receiving medical attention. Using items previously used to assess prevalence of health behaviour,47–49 clients were asked to report whether they were a smoker of any tobacco products,47 how many serves of vegetables and fruit they usually consumed each day48 and how many days a week they usually undertook 30 min or more of physical activity.49 Following Australian National Guidelines,36–39 participants were defined as being at risk if they: smoked any tobacco products,39 consumed less than two serves of fruit and/or five serves of vegetables per day37 or participated in less than 30 min of physical activity at least five times a week.38

During the clinician CATI, clinicians were asked to report their age, Aboriginal and/or Torres Strait Islander status, years working in community health and their current employment status. Clinician gender, discipline and service team were extracted from the electronic records system.

Client reported provision of preventive care

For assessment, clients were asked if a clinician asked about their smoking status, fruit and/or vegetable consumption and levels of physical activity (yes, no or don’t know). For provision of brief advice, clients classified as ‘at risk’ were asked if a clinician advised them to quit smoking or use nicotine replacement therapy, to increase their fruit and/or vegetable consumption or to participate in more physical activity (yes, no or don’t know). For referral, ‘at risk’ clients were asked if a clinician spoke to them about, or offered them a referral (yes, no or don’t know) to the NSW Quitline (for smoking)50 or the Get Healthy Information and Coaching Service (for physical activity and nutrition).51 Clients were also asked if a clinician recommended any other supports (eg, GP, AMS, dietitian, internet websites, physical activity classes and physiotherapist) (yes, no or don’t know).

Clinician reported provision of preventive care

On a scale of 0%–100% or ‘don’t know’,52 53 clinicians were asked to report the proportion of clients in the previous 2 months they asked about each of the health risk behaviours (assessment), the proportion of ‘at risk’ clients for which they advised to modify their risky health behaviour(s) (brief advice) and the proportion of ‘at risk’ clients they provided with referral options. Referral measures were the same as those for client report.

Client reported acceptability of preventive care provision

At follow-up, clients were asked to indicate their level of agreement (strongly agree, agree, unsure, disagree or strongly disagree) with three statements for each behavioural risk regarding: the acceptability of substance use treatment clinicians asking about their health behaviours, and for clients at risk, providing brief advice and arranging a referral to further support.

Clinician reported attitudes regarding the provision of preventive care

At follow-up, clinicians were asked to indicate their level of agreement (strongly agree, agree, unsure, disagree, strongly disagree or refused) with 13 statements within three attitudinal domains: perceived role in the provision of preventive care; self-efficacy in providing preventive care; and perception of client interest in modifying health risk behaviours.54

Statistical power

Client report

Of the 30 clients selected each week in the baseline and follow-up periods, it was estimated that 75% would be eligible for contact (n=23), 60% of eligible clients would be contactable (n=14) and of those contacted 70% would complete the survey (n=9).55 Therefore, it was expected that 234 clients would participate in the CATI at baseline and at follow-up. Assuming a baseline prevalence of care of 23%,29 it was estimated the study would have 80% power to detect an increase in assessment of 15% from baseline to follow-up (α=0.01). Using the least prevalent health risk behaviour (physical activity; 31%56) and an assumed prevalence of care at baseline of 25%,29 the study was estimated to have 80% power to detect an increase of 27% in the provision of brief advice and offer of referral.

Statistical analysis

Analyses were undertaken using SAS V.9.3. Client residential postcodes were used to determine disadvantage (Socio-Economic Indexes for Areas (SEIFA)) and remoteness (Access/Remoteness Index of Australia). SEIFA codes were collapsed into low, medium and high disadvantage, and geographic location collapsed into major cities versus regional/remote towns.57 58 Client reported provision of each element of care was dichotomised into ‘yes’ and ‘no’ (no or don’t know) responses. Client reported acceptability of care and clinician reported attitudes regarding preventive care were dichotomised into ‘agree’ (strongly agree oragree) and ‘disagree’ (unsure, disagree or strongly disagree).

An additional variable, provision of ‘care for all risks’,52 53 was created for the proportion of clients who: were assessed for all risk behaviours (assessment); given brief advice for all their risk behaviours (brief advice); and offered a referral for all their risk behaviours (offer to arrange referral to telephone helplines).

Based on recommendations of clinical guidelines19–21 and consistent with prior studies,52 53 clinician reported optimal care was defined as care provided to 80% or more of clients and was considered an appropriate benchmark for care provision. In addition, an overall ‘optimal care’ variable was created for each element of care (assessment, brief advice, spoke about telephone service, arrange referral to telephone service, advise GP/AMS and advise other support) for each risk behaviour separately and for all behaviours combined.52 53 59

χ2 analyses were used to compare baseline and follow-up characteristics of clients and clinicians and characteristics between participants and non-participants.

Provision of preventive care

For client report, logistic regression was used to examine change in care provision between baseline and follow-up. Analyses were undertaken for each element of care for the three health risks behaviours individually and for ‘care for all risks’ (20 models). All models were adjusted for age, gender, service type and number of appointments.

For clinician report, χ2 analyses were used to compare the prevalence of ‘optimal’ preventive care at baseline and follow-up for each care element for each health risk behaviour individually and for all health risk behaviours combined (24 outcomes).

A Bonferroni adjusted significance level of p<0.003 was used for all care provision analyses.

Results

Client sample

Of the 1132 clients randomly selected to participate in the surveys, 306 (27%) were unable to be contacted to complete the survey and 125 (11%) were ineligible. Of the 701 contactable and eligible clients, 415 (59%) agreed to participate. Compared with non-participants, participants were less likely to be of Aboriginal and/or Torres Strait Islander origin (28% vs 15%, p<0.001) or be under 40 years of age (67% vs 51%, p<0.001) and more likely to have had only one service appointment (58% vs 64%, p<0.001). Compared with baseline, clients at follow-up were more likely to have completed a technical certificate, university or higher degree and to have had one service appointment (table 1).

Table 1

Client characteristics (baseline n=226, follow-up n=189)

Characteristic	Baseline n (%)	Follow-up n (%)	P values
Female	70 (31)	52 (28)	0.44
Aboriginal and/or Torres Strait Islander	50 (22)	33 (18)	0.11
Age (years)			0.44
18–34	89 (39)	67 (35)
35–54	113 (50)	106 (56)
55+	24 (11)	16 (8)
Employment status			0.05
Employed	65 (29)	72 (38)
Not employed	161 (71)	117 (62)
Marital status
Living with partner	158 (70)	123 (65)	0.29
Not living with partner	68 (30)	66 (35)
Highest education level completed			0.006
High school or less	49 (22)	34 (18)
Completed high school	95 (42)	57 (30)
Technical certificate, university degree or higher	82 (36)	98 (52)
Geographic location			0.07
Major cities	67 (30)	41 (22)
Regional/remote	159 (70)	148 (78)
Index of disadvantage			0.57
Low	13 (6)	9 (5)
Moderate	104 (46)	94 (5)
High	108 (48)	86 (45)
Service type			0.08
Counselling	182 (81)	167 (89)
Stimulant treatment	14 (6)	7 (4)
Court diversion	29 (13)	14 (7)
Times seen in the last 12 months
Mean (SD)	3.4 (5.1)	1.6 (1.5)
1	106 (47)	131 (69)	<0.001
2–4	80 (35)	52 (28)
5–11	27 (12)	5 (3)
12+	13 (6)	1 (1)
Risk factor
Smoking	174 (77)	129 (68)	0.04
Insufficient fruit and/or vegetable consumption	196 (87)	150 (79)	0.03
Insufficient physical activity	65 (29)	67 (36)	0.15

Clinician sample

Of the 186 clinicians invited to participate in the surveys, 146 (78%) were eligible of whom 100 (68%) agreed to participate. Compared with baseline, participating clinicians at follow-up were more likely to work in their discipline for 10 years or more (table 2).

Table 2

Clinician characteristics

Characteristic	Baseline (n=54) n (%)	Follow-up (n=46) n (%)	P values
Female	40 (74)	36 (78)	0.63
Aboriginal and/or Torres Strait Islander origin	7 (13)	2 (4)	0.12
Age (years)			0.79
20–39	12 (22)	11 (24)
40+	42 (78)	34 (76)
Years in community health			0.87
<2	7 (13)	6 (13)
3–9	14 (26)	14 (30)
10+	33 (61)	26 (57)
Discipline			0.77
Nurse	25 (46)	22 (48)
Psychologist/  counsellor/  social worker	23 (43)	17 (37)
Other	6 (11)	7 (15)
Years in discipline			<0.001
<2	16 (30)	2 (4)
3–4	17 (32)	10 (22)
10+	21 (39)	34 (74)
Service team			0.52
Counselling	18 (33)	18 (39)
Pharmacotherapy	13 (24)	13 (28)
Stimulant treatment	15 (28)	7 (15)
Court diversion programs	8 (15)	8 (17)
Employment status			0.58
Full time	37 (69)	35 (76)
Part time	14 (26)	10 (22)
Casual	3 (6)	1 (2)

Provision of preventive care increased from baseline to follow-up for 6 out of 20 measures. Assessment of health behaviour risks increased for insufficient fruit and/or vegetable consumption (24% vs 54%, p<0.001) and for all risks combined (18% vs 48%, p<0.001). Brief advice increased for insufficient fruit and/or vegetable consumption (26% vs 46%, p<0.001) and all risks combined (26% vs 44%, p<0.001). Clinicians speaking about the helpline increased for insufficient fruit and/or vegetable consumption (10% vs 31%, p<0.001) and for all risks combined (11% vs 28%, p<0.001) (table 3). No increases were indicated any form of preventive care for smoking or insufficient physical activity.

Table 3

Comparison of client reported provision of preventive care between baseline and follow-up (baseline: n=226, follow-up: n=189)

Outcome	Baseline n (%)	Follow-up n (%)	OR (95% CI)	P values
Assessment
Smoking	202 (90)	176 (93)	1.6 (0.7 to 3.2)	0.24
Insufficient fruit and/or vegetable consumption	55 (24)	102 (54)	3.4 (2.2 to 5.3)	<0.001
Insufficient physical activity	128 (57)	122 (65)	1.5 (1.0 to 2.3)	0.07
All risks	42 (19)	90 (48)	3.8 (2.4 to 6.0)	<0.001
Brief advice*
Smoking	134 (77)	88 (68)	0.6 (0.3 to 1.0)	0.05
Insufficient fruit and/or vegetable consumption	50 (26)	69 (46)	2.5 (1.6 to 4.1)	<0.001
Insufficient physical activity	35 (54)	40 (60)	1.7 (0.9 to 3.9)	0.17
All risks	57 (26)	79 (44)	2.4 (1.5 to 3.8)	<0.001
Spoke about telephone helplines*
Smoking (Quitline)	87 (50)	64 (50)	1.0 (0.6 to 1.6)	0.96
Insufficient fruit and/or vegetable consumption (Get Healthy)	19 (10)	47 (31)	4.1 (2.2 to 7.7)	<0.001
Insufficient physical activity (Get Healthy)	6 (9)	17 (25)	2.4 (0.8 to 7.0)	0.11
All risks	24 (11)	50 (28)	3.2 (1.8 to 5.6)	<0.001
Offered to arrange a referral to telephone helplines*
Smoking (Quitline)	11 (6)	13 (10)	2.7 (1.0 to 7.2)	0.05
Insufficient fruit and/or vegetable consumption (Get Healthy)	1 (1)	12 (8)	18.4 (2.3 to 149.8)	0.006
Insufficient physical activity (Get Healthy)	2 (3)	3 (5)	1.4 (0.2 to 9.8)	0.75
All risks	2 (1)	9 (5)	9.1 (1.5 to 56.7)	0.02
Recommend other support (includes GP/AMS)*†
Smoking	47 (27)	31 (24)	0.9 (0.5 to 1.7)	0.85
Insufficient fruit and/or vegetable consumption	19 (10)	16 (11)	3.5 (1.2 to 9.8)	0.94
Insufficient physical activity	13 (20)	21 (31)	1.0 (0.5 to 2.1)	0.02
All risks	11 (5)	15 (8)	1.7 (0.7 to 4.1)	0.20

*Sample only include clients who had reported not meeting the guidelines for the relevant health risk behaviour. Smoking: baseline=174, follow-up=129; insufficient fruit and/or vegetable consumption: baseline=196, follow-up=150; insufficient physical activity: baseline=65, follow-up=67 (table 1).

†Other support included: GP, AMS dietitian, internet websites, physical activity classes and physiotherapist.

AMS, Aboriginal Medical Service; GP, general practitioner.

Clinician-reported provision of optimal preventive care

Provision of optimal care increased from baseline to follow-up for 6 out of 24 measures including: assessment of insufficient fruit and/or vegetable consumption (22% vs 63%, p<0.001) and all risks combined (15% vs 61%, p<0.001), and talking to clients about the telephone helplines for smoking (41% vs 74%, p=0.003), insufficient fruit and/or vegetable consumption (11% vs 57%, p<0.001), insufficient physical activity (9% vs 41%, p<0.001) and all risks combined (6% vs 30%, p<0.001) (table 4).

Table 4

Clinician reported provision of ‘optimal’ care (provided to >80% clients)

Outcome	Baseline n=54 n (%)	Follow-up n=46 n (%)	P values
Assessment
Smoking	47 (87)	46 (100)	0.01
Insufficient fruit and/or vegetable consumption	12 (22)	29 (63)	<0.001
Insufficient physical activity	24 (44)	32 (70)	0.01
All risks	8 (15)	28 (61)	<0.001
Brief advice
Smoking (quit/NRT)	43 (80)	40 (87)	0.33
Insufficient fruit and/or vegetable consumption	26 (48)	31 (67)	0.05
Insufficient physical activity	28 (52)	24 (52)	0.97
All risks	33 (61)	27 (59)	0.81
Referral Spoke about telephone service
Smoking (Quitline)	22 (41)	34 (74)	<0.001
Insufficient fruit and/or vegetable consumption (Get Healthy)	6 (11)	26 (57)	<0.001
Insufficient physical activity (Get Healthy)	5 (9)	19 (41)	<0.001
All risks	3 (6)	14 (30)	<0.001
Arranged referral to telephone service
Smoking (Quitline)	0 (0)	2 (4)	0.12
Insufficient fruit and/or vegetable consumption (Get Healthy)	2 (4)	1 (2)	0.65
Insufficient physical activity (Get Healthy)	1 (2)	1 (2)	0.91
All risks	0 (0)	1 (2)	0.28
Advised GP/AMS
Smoking	21 (39)	18 (39)	0.98
Insufficient fruit and/or vegetable consumption	6 (11)	11 (24)	0.09
Insufficient physical activity	4 (7)	6 (13)	0.35
All risks	3 (6)	4 (9)	0.54
Advised other types of support*
Smoking	18 (33)	14 (30)	0.76
Insufficient fruit and/or vegetable consumption	7 (13)	10 (22)	0.24
Insufficient physical activity	11 (20)	11 (24)	0.67
All risks	4 (7)	2 (4)	0.52

*Other support included: dietitian, internet websites, physical activity classes and physiotherapist.

AMS, Aboriginal Medical Service; GP, general practitioner; NRT, nicotine replacement therapy.

Client acceptability of preventive care

Across all risk behaviours and each element of care, the majority of clients agreed that such care was acceptable (85%–99%).

Clinician-reported attitudes regarding the provision of preventive care

Overall, the majority of clinicians reported high levels of agreement with the attitudinal statements (>75% agreement with 9 of 12 statements). The least supported statements were from each of the three attitudinal domains: providing preventive care leaves time to undertake acute care (65%); clients will change all their health behaviours due to the care provided (57%); and clients were interested in changing their health behaviours (39%) (table 5).

Table 5

Clinician reported attitudes regarding preventive care at follow-up (agree/strongly agree)

Attitudinal item	n (%)
Role congruence
It is part of my role as a community health clinician to provide preventive care to clients regarding their smoking behaviours.	45 (98)
My manager believes the provision of preventive care is important.	43 (93)
It is part of my role as a community health clinician to provide preventive care to clients regarding their physical activity levels.	39 (85)
It is part of my role as a community health clinician to provide preventive care to clients regarding their fruit and vegetable consumption.	37 (80)
Addressing health risk behaviours with clients does not jeopardise my relationship with the client.*	36 (78)
Providing preventive care for health risk behaviours leaves me time to undertake acute care of the client.*	30 (65)
Self-efficacy
I have the knowledge and skills to provide preventive care to clients regarding all health risk behaviours.	46 (100)
I feel confident to talk with clients about all their health risk behaviours.	45 (97)
There are services I can refer my clients to that provide support to change all their health risk behaviours.	43 (93)
Clients find it acceptable for me to talk with them about their health risk behaviours.	37 (80)
Clients will change their health risk behaviours because of the care I can provide them.	26 (56)
Perceived client interest
Clients I see are not generally too old to benefit from changing their health risk behaviours.	46 (100)
Clients I see are interested in changing their health risk behaviours.*	18 (39)

*Item originally worded as a negative statement and has been reversed.

Discussion

This study investigated the potential effectiveness of a clinical practice change intervention in increasing clinician provision of preventive care for multiple health risk behaviours within community-based substance use services. Across four behaviours, three elements of preventive care and client and clinician report, significant increases in care provision were observed for 12 of a total of 44 outcomes assessed, primarily for insufficient fruit and vegetable consumption. The increased care provision found for such outcomes suggests the intervention has the potential to increase preventive care delivery.

The greatest intervention effect was indicated for insufficient fruit and/or vegetable consumption for both client and clinician report and suggests clinicians can be assisted to provide preventive care addressing such risks within community substance use treatment settings. The importance of a strategy to address insufficient fruit and/or vegetable consumption is emphasised by the significant health burden associated with such risks,60 61 the high prevalence of such risks within substance use treatment clients56 and the low levels of care being provided at baseline.

The only increase in provision of smoking-related care was found for speaking about the Quitline to clients who were smokers, as reported by clinicians but not clients. Provision of assessment and brief advice for smoking may not have increased due to the high levels of provision at baseline: 87%–90% and 77%–80%, respectively. Such high levels of care may be due to clinicians feeling more confident to address smoking given it is the risk behaviour most closely aligned with substance use treatment services.12

Increases in provision of care for insufficient physical activity were only observed for clinician reported speaking about the Get Healthy Information and Coaching helpline. The prevalence of insufficient physical activity compared with the other health risk behaviours was lower and therefore addressing such risk may not be perceived as a treatment priority. In addition, although clinicians may recognise the importance of assessing physical activity for all clients, to overcome barriers such as competing clinical priorities, clinicians may target high-risk clients for care provision around insufficient physical activity.62

Little intervention impact was indicated for referral to further specialist behaviour change support across all behaviours and referral to the helplines remained low (<10%). Low levels of arranging referral may be due to clients declining an offer of referral or clinicians delaying referral due to clients’ acute substance use problems.63 However, given the high level of client acceptability regarding referral (85%–95%), other barriers to the routine referral of clients are likely.

One such potential barrier is clinician belief that clients are not interested in improving their health risk behaviours. Only 39% of clinicians reported their clients were interested in modifying their health behaviours. However, research indicates that substance use service clients are interested in quitting smoking (67%–69%),14 56 increasing fruit and/or vegetable consumption (55%)56 and increasing physical activity (67%–81%).14 56 In addition, over a half of clinicians (56%) believed that clients would change all their health risk behaviours due to the care provided. Such beliefs may be influenced by the clients’ other presenting issues, such as legal or child protections problems or partner violence.64 A further barrier to preventive care delivery may be the lack of formal guidelines for preventive care that recognise the unique aspects of substance use treatment.65

The findings of this study should be considered in light of its strengths and limitations. A strength of this study is the use of both client and clinician report. Although there is no gold standard for measuring preventive care provision,66 multiple data collection methods can demonstrate consistent trends67 such as that found regarding fruit and/or vegetable consumption and referral. However, inconsistencies between client and clinician report were also observed. Both measurement approaches have inherent potential bias, and understanding which approach represents reality is difficulty beyond direct observation.68 69 However, client recall of preventive care may be judged a more important indicator of care delivery, given clients must recall receiving advice in order to prompt health risk behaviour change.67 The primary study limitation is the pre-post design, although it was considered appropriate for this preliminary examination of the potential effectiveness of a clinician practice change intervention. Furthermore, due to the lack of research regarding preventive care within substance use treatment settings, evidence for the use of the 2As+R model was derived from general healthcare settings. However, this model may not be the most appropriate model of preventive care for this setting and warrants further investigation. A further limitation is that the study was conducted in one district within Australia and therefore the generalisability of findings to other services is unknown. In addition, the generalisability of the results may be limited by the differences in demographics between the baseline and follow-up clinician and client samples.

To the authors’ knowledge this is the first study to evaluate the effectiveness of a multistrategic clinical practice intervention on the provision of preventive care to community-based substance use clients for multiple health risk behaviours. The outcomes of this pilot study can be used to inform future, more rigorously designed, interventions. Future interventions can be improved by further investigating the barriers to the provision and uptake of preventive care in substance use settings and selecting strategies to address such barriers.