Globally, harmful alcohol use is a causal factor in more than 200 diseases and injuries and is responsible for 3 million deaths every year (WHO, 2022). One approach to reducing the burden of harmful alcohol use is to implement population-based screening and brief intervention programmes focused on preventing unhealthy alcohol use.
Primary care interventions represent a frontline opportunity to identify risks and provide brief advice. Supporting evidence suggests that primary care brief intervention trials significantly reduce alcohol consumption (Beyer et al., 2018). Similarly, evidence from randomised controlled trials has shown primary care management of Alcohol Use Disorder (AUD) is effective at reducing heavy drinking (Anton et al, 2006).
However, whilst primary care settings offer the opportunity for low-cost alcohol screening and brief intervention and treatment of AUD, they are scarcely adopted in practice. Despite a strong evidence base for population-based primary care prevention and treatment of alcohol-related problems, many individuals at-risk from unhealthy and harmful alcohol use do not receive the support they need.
This study (Lee et al., 2023) reports on the Sustained Patient-Centred Alcohol-Related Care (SPARC) trial, a randomised implementation trial aimed at improving population-level alcohol harm prevention and AUD treatment across primary care practices in the US.
The trial aimed to investigate where the SPARC intervention, compared to usual care, increased the proportion of primary care patients who screened positive for harmful alcohol use and received a brief intervention. Secondly, the trial examined whether the SPARC intervention increased the proportion of primary care patients who were newly diagnosed with AUD and received AUD treatment.
In a stepped-wedge, cluster randomised implementation trial, 25 primary care practices in Washington State, US (with no prior population-based alcohol screening, brief intervention, and AUD diagnosis and treatment) were included.
Implementation of the SPARC intervention was part of a broader programme of behavioural health including screening and addressing depression, suicidality and other drug use. Each practice in the study was randomly assigned a start date, with study waves staggered by 4 months. Months before the trial were labelled “usual care” and those after “SPARC intervention” periods.
All patients were at least 18 years old with a visit to primary care between January 1st, 2015, and July 31st, 2018, and data were extracted from Electronic Health Records (EHR) and insurance claims.
The SPARC intervention was designed to implement population-based approach:
- Alcohol screening with the Alcohol Use Disorders Identification Test Consumption (AUDIT-C)
- Brief intervention for patients who screened positive (AUDIT-C ≥ 3 for females and ≥ 4 for males)
- Assessment with an alcohol symptom checklist for (DSM-5) AUD for patients with high-risk drinking AUDIT-C ≥ 7
- Shared decision-making about AUD treatment options
- Support for initiation and engagement in AUD treatment.
The intervention used three evidence-based implementation strategies:
- Performance feedback (weekly reporting on the prevalence and assessment of AUD symptoms)
- Practice facilitators (expertise in alcohol-related care, addressed stigma, encouraged patient-centred decision-making for AUD, training on behavioural health integration)
- EHR clinical decision support (EHR prompts for screening assessment for AUD symptoms and treatment initiation).
The primary outcome for prevention was the presence of both a positive alcohol screen on the day of a primary care visit or during the past year and EHR documentation of brief intervention in the next 14 days. Brief intervention was indicated by 1 or more EHR records (including diagnosis codes for brief interventions, brief intervention templates and orders for leaflets on unhealthy alcohol use).
The primary outcome for AUD treatment engagement was both a new AUD diagnosis on the day of a primary care visit (with no diagnosis in the previous 365 days) and follow-up in-person treatment initiation and engagement. Initiation required an ICD code within 14 days after AUD diagnoses, and engagement required 2 or more visits with ICD codes within 30 days of initiation.
Secondary outcomes included AUDIT-C positive screens, high positive screens that prompted assessment with Alcohol Symptom Checklist, completion of Alcohol Symptom Checklist, new AUD diagnosis and AUD treatment initiation.
- Participants were 333,596 primary care patients, with a mean age of 48 years and predominantly female (58%) and white (70%)
- 255,789 patients were seen during usual care versus 228,258 during the SPARC trial.
Prevention and brief intervention
- The proportion of patients with a brief intervention in the EHR (Electronic Health Record) was greater during the SPARC trial compared with usual care (p < 0.001)
- The proportion of patients with documented alcohol screening (p < 0.001) and a positive alcohol screen (p < 0.001) were also greater during the SPARC trial compared with usual care.
AUD treatment initiation and engagement
- The proportion of patients with AUD engagement did not differ between the SPARC trial and usual care periods (p = 0.30)
- The proportion of patients assessed with an Alcohol Symptom Checklist (p < 0.001), who had a new AUD diagnosis documented (p = 0.003) and initiation of AUD treatment (p = 0.04) was greater during the SPARC trial compared to usual care.
Findings indicate that a population-based primary care trial increased the prevention of unhealthy alcohol use via mechanisms of increased alcohol screening and positive alcohol screens and a greater proportion of brief interventions.
The trial did not increase AUD treatment engagement, but did increase key markers in the AUD treatment pathway, including new AUD diagnoses and initiations of AUD treatment.
Strengths and limitations
SPARC is the first implementation trial to increase brief intervention, AUD diagnosis and initiation of AUD treatment in primary care without the use of research-supported clinicians.
These findings demonstrate that population-based implementation trials can improve prevention and treatment outcomes for harmful alcohol use outside of the closely controlled parameters of clinical trial settings. However, the magnitude of the increases in brief interventions, AUD diagnosis and initiation of treatment are modest. This finding is consistent with other implementation trials in primary care settings that report modest prevention and treatment effects. It also remains unknown whether these effects are sustained over time.
The SPARC trial goes beyond the typical use of EHR as an implementation strategy to include performance feedback and practice facilitation. These tools have been shown to be effective in supporting primary care clinicians to include screening and brief intervention in their practice (Fleming et al, 2004). However, given the wide range of intervention elements and implementation structures used in the present, it is difficult to determine which are most effective in supporting the prevention and treatment of AUD.
Implications for practice
Despite the presented evidence for low-cost, population-based implementation of prevention of harmful alcohol use and treatment of AUD, these practices are rarely used in primary care. In a UK-based survey, only one in ten heavy drinkers reported receiving brief advice about their drinking in a primary care setting (Brown et al, 2016).
Therefore, the success of screening, brief intervention, and initiation of AUD treatment is dependent on their utilisation in primary care practice. Several studies have reported on the challenges facing primary care clinicians in implementing prevention and treatment approaches including, lack of training (Aira et al, 2003) and emotional and moral difficulties in raising and assessing alcohol-related problems in consultations (Rapley et al, 2006).
The present findings add to a growing body of evidence reporting modest effects of primary care-based prevention and treatment of AUD, suggesting ongoing difficulties with implementation. Therefore, it is recommended that future research focuses on identifying facilitators for implementation, including the type of implementation strategy, and length and magnitude of intervention effectiveness.
Statement of interest
SA conducts research in the area of alcohol and substance misuse. No conflicts of interest with the paper reviewed in this blog.
Lee AKIntegrating Alcohol-Related Prevention and Treatment Into Primary Care: A Cluster Randomized Implementation Trial. JAMA Intern Med. 2023;183(4):319–328. doi:10.1001/jamainternmed.2022.7083Bobb JF Richards JE, et al.
Aira M, Kauhanen J, Larivaara P, Rautio P. Factors influencing inquiry about patients’ alcohol consumption by primary health care physicians: qualitative semi-structured interview study. Fam Pract. 2003 Jun;20(3):270-5. doi: 10.1093/fampra/cmg307.
Anton RF, O’Malley SS, Ciraulo DA, Cisler RA, Couper D, Donovan DM, Gastfriend DR, Hosking JD, Johnson BA, LoCastro JS, Longabaugh R, Mason BJ, Mattson ME, Miller WR, Pettinati HM, Randall CL, Swift R, Weiss RD, Williams LD, Zweben A; COMBINE Study Research Group. Combined pharmacotherapies and behavioral interventions for alcohol dependence: the COMBINE study: a randomized controlled trial. JAMA. 2006 May 3;295(17):2003-17. doi: 10.1001/jama.295.17.2003.
Brown J, West R, Angus C, Beard E, Brennan A, Drummond C, Hickman M, Holmes J, Kaner E, Michie S. Comparison of brief interventions in primary care on smoking and excessive alcohol consumption: a population survey in England. Br J Gen Pract. 2016 Jan;66(642):e1-9. doi: 10.3399/bjgp16X683149.
Beyer F, Lynch E, Kaner E. Brief Interventions in Primary Care: an Evidence Overview of Practitioner and Digital Intervention Programmes. Curr Addict Rep. 2018;5(2):265-273. doi: 10.1007/s40429-018-0198-7. Epub 2018 May 3.
Fleming MF. Screening and brief intervention in primary care settings. Alcohol Res Health. 2004-2005;28(2):57-62.
Rapley T, May C, Frances Kaner E. Still a difficult business? Negotiating alcohol-related problems in general practice consultations. Soc Sci Med. 2006 Nov;63(9):2418-28. doi: 10.1016/j.socscimed.2006.05.025.
World Health Organisation (2022) https://www.who.int/news-room/fact-sheets/detail/alcohol