Aboriginal and Torres Strait Islander health services provide strengths- based, integrated programs to improve health and wellbeing outcomes for Aboriginal and Torres Strait Islander people (hereafter, respectfully referred to collectively as Indigenous Australians; Appendix 1, available online only). However, access to culturally responsive health services, particularly medication management services, is not equitable.1–3 The national network of >5700 community pharmacies (CPs) presents an opportunity to improve accessibility to medication management services for Indigenous Australians.
Australian pharmacist-led medication management services (MedsCheck/Diabetes MedsCheck and Home Medicines Review) are government funded to address medication safety (Appendix 1). Indigenous Australians who face greater chronic illness burdens than non-Indigenous Australians4 and a range of medication-related problems (MRPs) could benefit substantially from these medication management services.5 Despite evidence that medication-related harm can cause preventable hospitalisations,6 barriers to medication management services for Indigenous Australians include cultural safety, geographical isolation, inflexible program rules and lower trust in non-Indigenous health services.1–3 Strategies to alleviate barriers include involving Aboriginal health workers (AHWs)/other support people and/or interpreters to ensure cultural and language translation; pharmacist training and mentoring; and funding models that facilitate flexible service delivery, encourage interprofessional relationships and address local needs.1–3
A culturally responsive medication management review service (the IMeRSe), delivered through community pharmacies and Aboriginal and Torres Strait Islander health services on a fee-for-service basis, was proposed to address barriers and improve access, using the existing pharmacist and AHW workforce, particularly in rural/remote regions. The aim of this study was to investigate the effectiveness of the IMeRSe service.
Methods
This pre-post single-arm study was conducted across nine Aboriginal and Torres Strait Islander health services and 23 associated community pharmacies from three urban, two rural and four remote locations in Queensland, New South Wales and the Northern Territory between June 2018 and October 2019 (Appendix 2, available online only). The study was prospectively registered with the Australian and New Zealand Clinical Trials Registry (ACTRN12618000188235), the protocol was published,7 and conduct and reporting were guided by the Consolidated Standards of Reporting Trials (CONSORT) statement.8 Ethics approval was provided by seven organisations (Appendix 3, available online only). Aboriginal and Torres Strait Islander health service CEO/Directors and pharmacy owners/managers provided written informed consent for their service/pharmacy to participate in the study. Consumers also provided written informed consent to participate in the study.
Recruitment of study sites and consumer participants
Eligibility criteria for sites, which were identified in consultation with study partners, included capacity to participate in training and facilitate IMeRSe delivery, to share electronic health records and work collaboratively. Engagement visits were conducted to eligible Aboriginal and Torres Strait Islander health service sites from May to November 2018, meeting staff, Elders, community boards/councils and pharmacists, and seeking input into local implementation. Study Champions and Study Coordinators were nominated.
Pharmacists completed a cross-cultural training program to build confidence and cultural capability for working effectively with Indigenous people and the Aboriginal and Torres Strait Islander health service.9 Key concepts introduced in an online module were re-inforced in a one-day interactive workshop and welcome/orientation sessions at local Aboriginal and Torres Strait Islander health services. Structured training was delivered to Aboriginal and Torres Strait Islander health service staff onsite regarding study processes. Pharmacists received ongoing support via webinars, with individual mentoring from a pharmacist mentor and a cultural mentor.9 Teleconferences and mentoring by an Indigenous researcher supported study coordinators.
Indigenous consumers were recruited from 1 June 2018 to 31 July 2019. Eligible consumers were adults (aged ≥18 years); living in the community with at least one chronic condition, and/or pregnant, and/or within 2 years postpartum; at risk of MRPs; engaged with the Aboriginal and Torres Strait Islander health service (three or more times in the last 2 years) and one of the associated pharmacies. Exclusion criterion was receiving a medication review (MedsCheck/Home Medicines Review [HMR]) within the previous 12 months.
IMeRSe process
This six-step, collaborative, medication management service involved consumer participants (and family members), pharmacists, AHWs and general practitioners (GPs), with follow-up(s) as required over 6 months (Box 1). The six steps comprised: (i) consumer recruitment and consent; (ii) collection of participant baseline data and referral for Medicines Talk; (iii) Medicines Talk with participant, pharmacist and AHW/support person; (iv) completion of Medicines Report by a pharmacist via an online system and GP notification; (v) GP response to pharmacist recommendations and completion of My Medicines Plan with the consumer and/or AHW; and (vi) structured clinical follow-up(s) and monitoring by a pharmacist with the participant and AHW/support person as required.
| Box 1. Overview of the six-step IMeRSe service |
|
Step one: Consumer recruitment and informed consent
- Study coordinators, Aboriginal and Torres Strait Islander health service and CP staff identified consumers who were at risk of an MRP opportunistically or via targeted strategies.
- Opportunistic recruitment occurred during scheduled/unscheduled appointments, at specialist clinics, outreach visits, community barbeques, when dispensing medication, and self-referral.
- Targeted identification occurred through review of clinical databases, participants accessing other services (eg dose administration aids, exercise programs) and via recall lists.
- Coordinators screened participants for eligibility, explained study processes, obtained consent to participate and to access Aboriginal and Torres Strait Islander health service, CP and hospital records.
- Separate consent was obtained to access health resource use data (eg from Medicare Benefits Schedule, Pharmaceutical Benefits Scheme and pathology).
- Participants’ GPs were notified when their patients were enrolled and again when the Medicines Report, prepared by the CPT, was ready for review.
|
|
Step two: Referral, information exchange and appointments
- The coordinator documented baseline information including demographics, medication‑related beliefs and adherence, psychological wellbeing and empowerment.
- The coordinator booked appointments with the CPT to conduct Medicines Talks, provided health summaries to the CPT and facilitated further information exchange as required.
|
|
Step three: Medicines Talk (Tier 1)
- Medicines Talks began as initial conversations between participants and CPTs, and Aboriginal and Torres Strait Islander health service staff/other support person(s) according to participant preference, at a mutually agreed location (eg Aboriginal and Torres Strait Islander health service, community centre, CP or other culturally safe location).
- The Medicines Talk aimed to establish and/or strengthen relationships using the Stay Strong Plan and to conduct a conversational review of medication(s), medication-related beliefs and adherence.18,19
- The Stay Strong Plan was developed for Indigenous Australians as a holistic, strengths‑based tool that positions health workers and patients as collaborative partners by integrating the principles of therapeutic problem-solving, motivational interviewing and goal setting. It induced confidence in CPTs to explore participants’ family connections, strengths, worries and health-related goals.18,19 Each party had the opportunity to share and learn from each other; selected problems were resolved through explanation/education.
|
|
Step four: Medicines Report
- CPTs recorded key information from the Medicines Talk in a secure, online, shared record (GuildCare NGTM) and submitted Medicines Reports as a series of recommendations to participants’ GPs for review.
- Automatic emails alerted GPs that a Medicines Report was completed. CPTs consulted with coordinators and Aboriginal and Torres Strait Islander health service staff as needed.
|
|
Step five: My Medicines Plan
- GPs reviewed and selected one of three options for each recommendation made by the pharmacist: (i) accepted; (ii) declined; or (iii) amended recommendations online/or in hardcopy.A This informed the My Medicines Plan prepared in collaboration with the participant and/or Aboriginal and Torres Strait Islander health service staff.
- The Medicines Plan was to be provided to the participant in printed format and filed in their health record.
|
|
Step six: Structured follow-up and monitoring over the next 6 months (Tier 2)
- This final step included actions agreed on by the participant, CPT and GP in the My Medicines Plan, and additional CPT follow-up and monitoring as required.
- Examples included: face-to-face meetings in the CP, Aboriginal and Torres Strait Islander health service or at home (participant preference); and via telephone.
- CPTs recorded follow-ups online via the GuildCare NGTM software and updated the Medicines Report with new recommendations, which treating GPs could incorporate into an updated My Medicines Plan.
|
AMedicines Plans were intended to be completed in the GuildCare NGTM software; however, internet and IT issues in some settings meant that paper-based recording was also used.
CP, community pharmacy; CPT, community pharmacist; GP, general practitioner; MRP, medication- related problem. |
Outcome data were collected at baseline and at 6 months’ follow-up. All clinical data related to IMeRSe (eg Medicines Talks, Medicines Reports) were collected in a purpose-built module within the cloud-based pharmacy software platform GuildCare NG™ (GuildLink Pty Ltd, Melbourne, Vic, Australia). Data included pharmacist-identified MRPs and recommendations, GP acceptance/amendments of recommendations and actions documented in Medicines Plans. Potential severity of pharmacist-identified MRPs was classified post-hoc by an independent assessor (experienced pharmacist accredited to conduct HMRs) using the validated Australian DOCUMENT framework.10
Outcomes
The primary outcome was the difference in cumulative incidence of serious MRPs in the 6 months before and after IMeRSe introduction. Serious MRPs were refined from existing literature11 by a Clinical Validation Group (CVG) of 13 clinical experts12–14 (Appendix 4, available online only). A serious MRP was defined as meeting the criteria of suboptimal patterns of care with/without causing a hospitalisation. Serious MRPs were identified from administrative data records (linked PBS and MBS data [Approval for Medicare Benefits Schedule [MBS] and Pharmaceutical Benefit Schedule [PBS] data was provided by the Services Australia External Request Evaluation Committee [EREC] of Australian Government [Ref: M19435; 20.04.18]]; Appendix 1).
Secondary outcomes were: potentially preventable medication-related hospitalisations (PPMRH), the subset of MRPs for which a hospitalisation is realised; medication adherence assessed using the Reported Adherence to Medication scale (RAM),15 Medication Possession Ratio (MPR)15 and Proportion of Days Covered (PDC)16 (Appendix 5, available online only); psychological and social empowerment (Growth and Empowerment Measure [GEM]17); psychological distress (Kessler Psychological Distress score [K10]18); beliefs about medicines (Beliefs about Medicines Questionnaire-specific scale [BMQ]15); treatment satisfaction and confidence (Treatment Satisfaction Questionnaire for Medication [TSQM]19); self-reported general health (Likert scale); classification of all MRPs (type and severity); and incidence of PPMRHs (Appendix 4). A medication adherence package (Medadhere) was used to estimate MPR and PDC; objective adherence measures using administrative data (PBS; Appendix 5).
Acceptability outcomes were: recruitment, drop-out and follow-up rates, and participant satisfaction with IMeRSe.
Statistical analysis
Sample size calculations were based on the primary outcome and assumed the serious MRP rate would decrease by 30% from 0.07 per person to 0.049 per person per 6 months, and with a standard deviation in the difference in serious MRPs of 0.14 (power=80%; α=0.05). With scant evidence from the literature available, our study sample size calculations were based on a previous study of HMR, which found a mean±standard deviation (SD) of 3.5±1.8 drug-related problems per patient reviewed.20 Given we did not know the comparability of our participants, we assumed 30% would have a mean of 3.5 problems and 70% a mean of 1.7 problems (1 SD lower), and that 3% would be considered serious MRPs.19 The assumption based on these calculations was that mean serious MRPs at baseline would be 0.07 per person. The 30% reduction in this rate was assumed based on clinical experience. Consequently, 6-month data were needed for 351 participants. Assuming 25% attrition, the recruitment target was 540.7
The primary outcome (serious MRPs) was analysed using a mixed-effects Poisson regression model. Time (pre/post) was entered as a fixed effect and participant as a random effect. The model was offset by the natural logarithm for each participant’s follow-up time, as some participants did not have full 6-month follow-up data. Effect estimates are reported as incidence rate ratios and 95% confidence intervals (CI). Secondary outcomes were compared between pre- and post-IMeRSe periods, using paired t-tests for continuous variables and McNemar’s test for paired nominal data. Data were analysed using Stata v13.0 (StataCorp, College Station, TX, USA). Statistical significance was P<0.05, with no adjustment for multiple comparisions.
The authors had full access to the data and analysis for this study. The data that support the findings of this study are available on request from the corresponding author, but restrictions apply to the availability of these data, which are not publicly available.
Results
Of the 311 consumers recruited, 291 were eligible and provided consent (Box 2). They were prescribed a median of seven (range 1–27) medications to treat a median of six (range 1–17) current health conditions, most commonly hypertension (77.3%), high cholesterol (76.3%) and diabetes (64.6%) (Table 1). The proportion of the IMeRSe cohort with major chronic health conditions was similar or higher than the general Indigenous population (Appendix 6, available online only), with heart disease, diabetes, kidney disease and stroke overrepresented.
Box 2. Overview of study enrolment, service delivery and follow-up.
ANot related to the IMeRSe service.
GP, general practitioner.
| Table 1. Consumer participant characteristics at baseline (n=291) |
| Characteristic |
Total
n=291
n (%) |
Medicines Talk
n=255
n (%) |
No intervention
n=36
n (%) |
P valueA |
| Age (years) |
|
|
|
0.51 |
| <45 |
24 (8.3) |
21 (8.2) |
3 (8.3) |
|
| 45–54 |
65 (22.3) |
58 (22.8) |
7 (19.4) |
|
| 55–64 |
95 (32.7) |
83 (32.6) |
12 (33.3) |
|
| 65–74 |
79 (27.2) |
66 (25.9) |
13 (36.1) |
|
| ≥75 |
28 (9.6) |
27 (10.6) |
1 (2.8) |
|
| Mean (SD) |
59.9 (11.6) |
60.0 (11.7) |
58.7 (11.1) |
0.52 |
| Gender |
|
|
|
0.15 |
| Male |
104 (35.7) |
95 (37.3) |
9 (25.0) |
|
| Female |
187 (64.3) |
160 (62.8) |
27 (75.0) |
|
| Cultural identity |
|
|
|
|
| Aboriginal |
267 (97.5) |
236 (97.1) |
31 |
|
| Torres Strait Islander |
2 (0.7) |
2 (0.8) |
Nil |
|
| Both |
5 (1.8) |
5 (2.1) |
Nil |
|
| Location |
|
|
|
0.02 |
| Urban |
74 (25.4) |
59 (23.1) |
15 (41.7) |
|
| Rural |
74 (25.4) |
70 (27.5) |
4 (11.1) |
|
| Remote |
143 (49.1) |
126 (49.4) |
17 (47.2) |
|
| Years with the Aboriginal and Torres Strait Islander health service |
|
|
|
1.00 |
| <5 |
39 (18.2) |
35 (18.3) |
4 (17.4) |
|
| ≥5 |
175 (81.8) |
156 (81.7) |
19 (82.6) |
|
| Self-reported general health |
|
|
|
0.68 |
| Excellent |
11 (4.0) |
10 (4.1) |
1 (3.2) |
|
| Very good |
47 (17.2) |
44 (18.1) |
3 (9.7) |
|
| Good |
126 (46.0) |
112 (46.1) |
14 (45.2) |
|
| Fair |
69 (25.2) |
59 (24.3) |
10 (32.3) |
|
| Poor |
21 (7.7) |
18 (7.4) |
3 (9.7) |
|
| Number of health conditions, median (IQR) |
6 (4) |
6 (4) |
6 (4) |
0.60B |
| DiabetesC |
188 (64.6) |
165 (64.7) |
23 (63.9) |
0.92 |
| High cholesterol |
222 (76.3) |
194 (76.1) |
28 (77.8) |
0.82 |
| Hypertension |
225 (77.3) |
194 (76.1) |
31 (86.1) |
0.18 |
| Ischaemic heart diseaseD |
82 (28.2) |
72 (28.2) |
10 (27.8) |
0.95 |
| Kidney disease |
68 (23.4) |
55 (21.6) |
13 (36.1) |
0.05 |
| Asthma |
65 (22.3) |
56 (22.0) |
9 (25.0) |
0.68 |
| Mental and behavioural problems |
90 (30.9) |
80 (31.4) |
10 (27.8) |
0.66 |
| Number of medications, median (IQR) |
7 (6) |
7 (6) |
6.5 (7) |
0.27B |
Note: Missing observations per characteristic: cultural identity, n=17; years with the Aboriginal and Torres Strait Islander health service, n=77; self- reported general health, n=17.
APearson’s chi-square test or Fisher’s exact test was used for categorical variables; one-way analysis of variance was used for continuous variable.
BTwo-sample Wilcoxon rank-sum test.
CIncluding type 1, type 2, unknown and gestational diabetes; the majority (94.2%) were type 2 diabetes.
DAngina, heart attack and other ischaemic heart disease.
IQR, interquartile range; SD, standard deviation. |
Overall, 255 (87.6%) completed a Medicines Talk with their pharmacist and had a Medicines Report provided to the Aboriginal and Torres Strait Islander health service and GP. Further, GPs finalised 240 (94.1%) management plans based on these reports. Medicines Talks took an average of 54.5 minutes; 54.1% occurred at the Aboriginal and Torres Strait Islander health service, 22.0% in CPs, 12.2% at community facilities and 11.8% in participants’ homes, with AHWs attending 49.0%. During Medicines Talks, pharmacists reviewed 2171 medicines (prescription n=2065; over-the-counter n=58; complementary n=38; bush medicines n=10) and identified 762 MRPs (range 0–10). Most MRPs related to undertreatment (24.8%), adherence issues (20.7%), education/information needs (12.3%), monitoring (12.2%) and drug selection issues (10.9%) (Appendix 7, available online only). Three MRPs were classified as high severity. Pharmacist recommendations mostly suggested review of/change in pharmacotherapy (44.4%; Appendix 7). GPs responded to 709 of 762 pharmacist recommendations, accepting 84.1%. Following Medicines Talks, 77 participants had 150 follow-up consultations.
For 207 (84.2%) participants with matched MBS/PBS data, mean (SD) incidence rate of serious MRPs per 6-months at risk reduced from 4.11 to 3.85 from pre- to post-IMeRSe (incidence rate-ratio [IRR]=0.93; 95% CI: 0.86–1.01; P=0.09; Table 2). Rate of PPMRHs pre- and post-IMeRSe was very similar (IRR=1.00; 95%CI 0.55–1.83; Table 2).
| Table 2. Incidence rate-ratio for serious medication-related problems and potentially preventable medication-related hospitalisations |
| Outcome |
nA |
N
pre |
N
post |
Pre-rate
(95% CI) |
Post-rate
(95% CI) |
Incidence rate-ratio (95% CI) |
P value |
| Serious MRPs |
207 |
1267 |
1184 |
4.11 (3.14–5.39) |
3.85 (2.94–5.04) |
0.93 (0.86–1.01) |
0.09 |
| PPMRHs |
207 |
21 |
21 |
0.06 (0.02–0.15) |
0.06 (0.02–0.15) |
1.00 (0.55–1.83) |
1.00 |
Note: The rate is reported as the rate per person per 6 months; pre- and post-rates and the incidence rate-ratio are estimated using a mixed-effects Poisson model. The primary outcome, serious MRPs, is significant at the P<0.10 level.
ANumber of participants with available Medicare Benefits Schedule and Pharmaceutical Benefits Scheme data for analysis.
CI, confidence interval; MRP, medication-related problem; PPMRH, potentially preventable medication-related hospitalisations. |
Positive differences between pre- and post-IMeRSe phases were found for participants’ concerns about their medicines (Beliefs about medicines questionnaire [BMQ]-concern scale); however, baseline beliefs about medicine necessity scores remained high (Table 3), indicating that participants believed that medicines were necessary for their health and wellbeing, and this did not really change post-IMeRSe. Self-reported medication adherence improved significantly (RAM scores, Table 3) as did objective adherence assessment (Appendix 5) and treatment satisfaction (TSQM scores, Table 3). Medication adherence, measured using the medication possession ratio (Appendix 5), increased significantly from median 0.83 (IQR 0.59, 0.99) to 0.92 (0.62, 1.02) (P=0.03).
| Table 3. Secondary outcomes (self-reported survey data) |
| |
nA |
Pre-score
mean (SD) |
Post-score
mean (SD) |
Mean difference
(95% CI) |
P value |
| BMQB |
| BMQ-necessity |
215 |
20.68 (3.62) |
20.56 (3.57) |
0.12 (–0.40 to 0.65) |
0.65 |
| BMQ-concern |
213 |
13.80 (4.54) |
13.00 (3.79) |
0.80 (0.18–1.42) |
0.01 |
| Adherence (RAM)C |
215 |
15.27 (3.29) |
15.84 (3.09) |
–0.57 (–1.04 to –0.09) |
0.02 |
| TSQMD |
| Effectiveness |
220 |
80.51 (20.70) |
86.74 (15.93) |
–6.24 (–9.24 to –3.23) |
<0.001 |
| Side effects |
10 |
81.88 (28.02) |
75.63 (31.52) |
6.25 (–21.23 to 33.73) |
0.620 |
| Convenience |
220 |
82.88 (22.76) |
89.72 (15.18) |
–6.84 (–9.73 to –3.96) |
<0.001 |
| Global satisfaction |
220 |
110.44 (30.50) |
118.87 (20.28) |
–8.43 (–12.48 to –4.38) |
<0.001 |
| GEME |
| GEM total score |
96 |
88.83 (16.93) |
92.96 (16.22) |
–4.13 (–7.19 to –1.06) |
0.009 |
| Inner peace |
96 |
31.50 (6.41) |
33.03 (6.09) |
–1.53 (–2.82 to –0.24) |
0.020 |
| Self-capacity |
96 |
20.10 (4.35) |
21.04 (4.22) |
–0.94 (–1.82 to –0.05) |
0.040 |
| Healing and growth |
96 |
21.33 (5.72) |
22.61 (4.81) |
–1.28 (–2.39 to –0.17) |
0.020 |
| Connection and purpose |
96 |
11.57 (2.73) |
11.95 (2.56) |
–0.38 (–0.95 to 0.20) |
0.200 |
| Psychological distress (K10)F |
| K10 total score |
94 |
19.04 (9.01) |
17.80 (7.95) |
1.24 (–0.47 to 2.96) |
0.15 |
| K10 distress |
|
Pre n (%) |
Post n (%) |
|
0.29 |
| Low (10–15) |
|
38/94 (40.04) |
41/94 (43.60) |
|
|
| Moderate (16–21) |
|
25/94 (26.60) |
28/94 (29.80) |
|
|
| High (22–29) |
|
16/94 (17.00) |
14/94 (14.90) |
|
|
| Very high (30–35) |
|
15/94 (16.00) |
11/94 (11.70) |
|
|
ANumber of participants with completed paired baseline and follow-up survey.
BHigher scores on specific-necessity scale indicate stronger beliefs that medicines are important to improve/maintain health and higher scores on specific concerns scale indicate higher concerns about taking medicines to maintain health.
CHigher scores indicate greater medication adherence.
DHigher scores indicate a higher level of treatment satisfaction.
EHigher scores indicate higher levels of empowerment (inner peace subscale items: EES 2 3 4 10 11 12 13 14; Self-capacity subscale items: EES 5 6 7 8 9; Healing and growth subscale items: scenarios 1 2 5 6; Connection and purpose subscale items: scenarios 3 4).
FLower scores indicate lower levels of psychological distress.
BMQ, Beliefs about Medicines Questionnaire; CI, confidence interval; EES, Emotional Empowerment Scale; GEM, Growth and Empowerment Measure; K10, Kessler psychological distress scale; RAM, Reported Adherence to Medication scale; SD, standard deviation; TSQM, Treatment Satisfaction Questionnaire for Medication. |
Participants rated their general health better post-IMeRSe (P=0.019, Wilcoxon signed-rank test; Appendix 8, available online only). High participant satisfaction for IMeRSe was reported; increased confidence and motivation to manage health and medication-related issues, and greater access to medicines via pharmacies (Figure 1; Appendix 8).
Figure 1. Participant satisfaction with the IMeRSe service.
Discussion
In this study, where community pharmacists worked with Aboriginal and Torres Strait Islander health service staff to deliver essential elements of a medication review as a culturally responsive service across diverse models of care and Australian Indigenous communities, the IMeRSe service was associated with decreased incidence of serious MRPs (modelled rate) by 7% at 6 months. The observed 7% reduction was lower than the anticipated reduction of 30% used in sample size calculations, and reduced participant recruitment potentially impacted study power. However, the observed baseline rate of serious MRPs (4.11 per 6-month period) was more than five-fold greater than anticipated (0.07 per 6-month period), suggesting that even a small relative reduction could lead to a clinically important absolute decrease of serious MRPs. Significant improvements observed in medication adherence (objective and subjective), concerns about medicines, treatment satisfaction and overall empowerment and wellbeing, point to longer-term health outcomes not captured within the 6-month observation period.
Other research has found that collaboration between GPs and pharmacists is significantly related to the implementation rate of medication review recommendations.21 IMeRSe facilitated a team-based approach between pharmacists and Aboriginal and Torres Strait Islander health service staff including GPs. Responsive GP input to reviewing pharmacists’ recommendations (94% of Medicines Reports reviewed with 84% recommendations accepted), particularly the 46% rated by the independent assessor as moderate severity and requiring GP treatment, were critical to improving participants’ health outcomes. The high level of engagement and collaboration observed in IMeRSe might be because of: team-based care being usual practice for GPs working in Aboriginal Health Services; and that working with local community pharmacists in the study who had pre-existing relationships with the consumer participants, GPs and AHWs promoted confidence in the medication review recommendations.
The flexibility of IMeRSe emphasised cultural safety with participant choice for location of Medicines Talks and AHW involvement. Most chose the Aboriginal and Torres Strait Islander health service for their appointment and approximately half chose to have an AHW alongside, re-inforcing the importance of trusting relationships and removing access barriers.
Another essential element supporting integrated IMeRSe delivery included secure transfer of health information between Aboriginal and Torres Strait Islander health service and CP staff via the GuildCare NG platform, although some Aboriginal and Torres Strait Islander health services preferred more traditional technologies (eg faxing).
Findings of the qualitative elements (participant and health practitioner experiences) and health expenditure will be published separately.
Key study strengths included use of administrative data to provide objective measures of medication safety (eg serious MRPs and PPMRHs) and triangulation of key outcomes and associated correlations (eg medication adherence/beliefs about medicines/treatment satisfaction). Limitations related to study design include the inability to attribute causal effects of the intervention; a quasi-experimental design compared outcomes observed pre- and post-IMeRSe. Selection bias might have been introduced by limitations related to study scope including site/consumer recruitment and research burden. Site recruitment was potentially limited by requirements for a minimum level of Aboriginal and Torres Strait Islander health service and pharmacy capacity to deliver IMeRSe, when local needs necessitated prioritisation of other activities, and concurrent recruitment to another Aboriginal and Torres Strait Islander health service-pharmacist trial. Restricted study timelines and research burden limited consumer recruitment, which potentially impacted study power.
Conclusion
This study demonstrates that a culturally responsive medication management service delivered by community pharmacists working with Aboriginal and Torres Strait Islander health service staff might reduce the incidence of serious MRPs and improve consumer health, medication use and wellbeing.
IMeRSe addressed established barriers to existing medication review programs with existing workforce and minimal changes to workflows to facilitate more equitable access for Indigenous Australians. GPs could consider including pharmacists regularly in team-based care such as in case conferences and using local community pharmacists to conduct culturally responsive medication reviews, which show promising engagement and outcomes for underserved populations.