Skip to main content

Table 2 Characteristics of studies included in the review

From: A systematic review of approaches to improve practice, detection and treatment of unhealthy alcohol use in primary health care: a role for continuous quality improvement

Study (n = 56)aSample sizeStrategyTargetsMain outcomesCQI elementsc
PatientsCliniciansSitesClin.actionbOrg. levelM/faceted123
Randomized controlled trials
WHO collaborative Project (Phase 3)
 Gomel 1998 AUS Alcohol [33]94,481Phase 1: 628 Phase 2: 161 (1 per site)628 161Phase 1: mailout (c); telemarketing; academic detailing. Phase 2: written guidance (c); training; training + min support; training + ongoing supportS, BI3,4YHigher uptake if academic detailing or telemarketing; higher screening in training or training + max support cf. other arms; advice significantly higher in max support armsNNN
Hansen et al. 1999 DNK Alcohol [34]na143naPhase 1 onlyS, BI4NHigher uptake if academic detailing or telemarketing. No significant differences control cf. intervention armsNNN
Kaner 1999 GBR Alcohol [39]11,007128128Phase 2 only (excluding the training + min support arm).S, BI4YIncreased implementation, screening and intervention in training + support.NNN
Funk 2005 AUS, BEL, DNK, NZ, ESP, GBR Alcohol [32]60,989Phase 1: 3436, Phase 2: 727 (1 per site)3436 727Phase 1 & 2 (excluding the training + min support arm)S, BI3,4YIncreased uptake if academic detailing or telemarketing. Increased screening and advice giving if training or training + support.NNN
Anderson 2004 AUS, BEL, ESP, GBR Alcohol [25]naPhase 1: 2924; Phase 2: 632 (1 per site)2924 632Phase 1 & 2 (excluding the training + min support arm)S, BI3,4YSub-analysis of Funk 2005: Increased screening and BI if physicians secure and committed in working with drinkersNNN
 Anderson 2016 ESP, GBR, NLD, POL, SWE Alcohol [23]Mean: 1500 consults/site (baseline)746120Country guidelines summary (c); training and support (TS), Financial reimbursement (FR), access to referral to eBI and combinations of theseS, BI3,4YDuring 12-week implementation: Increased screening in TS arm and FR arm. Increased intervention (screening or advice) in TS, FR, TSFR and TSFReBI. No effect on giving advice to screen positive pts.NNN
 Bendtsen 2016 ESP, GBR, NLD, POL, SWE Alcohol [26]As above35060As aboveS, BI3,4YNo association between eBI and increase in screening. Increased proportion of screen-positive pts. given BI. Low pt. and provider uptake rates of eBINNN
 Anderson 2017 ESP, GBR, NLD, POL, SWE Alcohol [24]As above746120As aboveS, BI3,4YAt 9 months follow-up: Increased intervention rates in TS. Reduction of intervention rates in all arms but reduction in TS arm was smaller.NNN
Chan 2013 AUS Broad prev [28].na1294Training (5As); Integrating assessment/prompts into initial visits; referral directory; resources including. Guides for nurses, action plans for each risk factor.S, BI, RT3,4YIncreased self-reported screening at 6 and 12 months (validated scale). No effect on self-reported management or referral.NNN
Harris 2013 AUS Broad prev [35].804na4As above.BI, RT3,4YIncrease in pt-reported referrals in intervention group at 3 months cf. baseline. No significant changes in self-reported alcohol consumption.N?N
Other RCTs
 Bonevski 1999 AUS Broad prev [27].291719naComputerized feedback system: guidelines, goal setting for GPs, GP feedback on performance in other health screening (not alcohol).na4YAlthough not targeted by intervention, at 3-month follow-up classification of hazardous/harmful drinkers more accurate in intervention arm cf. controls.NNN
 Dubey 2006 CAN Broad prev [30].1117384Gender based preventative checklist prompt with evidence-based recommendations.S4NSignificant increase in alcohol history intake between baseline and follow-up in intervention arm. Increase significantly associated with intervention.NNN
 Chossis 2007 Switzerland Alcohol [29]260272Training; summary checklist, textbook and pt. education materials.BI4YIntervention residents conducted more components of BI: more likely to explain safe drinking limits provide feedback, seek pt. opinions on drinking limits, after training but not at follow-up; no effect on pt. drinking patterns.NNN
 Friedmann 2006 USA Alcohol [31]164182Maintenance care training for alcohol problems in remission (5As); follow-up academic detailing; booster training; materials for pts. and clinicians; pt. record prompt (paper).S, BI, RP, RT4YIntervention pts. more likely to report clinician took alcohol history. Intervention clinicians who took alcohol history more likely to assess prior and planned alcohol treatment, offer of prescriptions and referral.NNN
 Harris 2015 AUS Broad prev [36].21,84812232Training of practice staff and QI facilitators; audit and feedback; site visits with goal setting; pt. education and referral materials; implementation support; facilitator support.S3,4YAt 12 months follow-up increase in odds of alcohol recording of alcohol consumption in the intervention compared to control. No significant change in the level of risk factors based on audit data.YYY
Haskard 2008 USA Broad prev [37].21961563Physician training; pt. training.BI4,5YSignificant upward trend in counselling to quit alcohol at time 6-months post training after initial expected drop at 1-month post training.NNN
Kaner 2003 GBR Alcohol [38]5541na212Phase 1: mailout (c); telemarketing; academic detailing. Phase 2: written guidance (c); training; training + ongoing support (directed at nurses).S, BI YIncreased implementation in training and training + support. Increased BI in training and training + support. Fewer pt. management errors in controls.N?N
Krist 2016 USA Broad prev [40].291315618MOHR: self-administered health behaviour questionnaire; MOHR summary and feedback for pts.; a summary of positive MOHR for clinicians; optional training for clinicians; freedom of method of implementation.S, BI, RT3,4YSignificantly higher screening for alcohol and goal setting to reduce risky drinking in intervention arm compared to control. No significant changes in referrals. No significant changes in alcohol consumption.NYN
Ornstein 2013 USA Alcohol [43]26,0057720Pre-intervention visit; electronic screening/intervention prompt and resource template; network meeting to discuss facilitators and barriers and to develop implementation plans; performance feedback; on-site support visits.S, BI, RP3,4YEarly Intervention (EI) phase: increased odds of screening and BI in EI cf. delayed intervention (DI); performance stable at DI phase. DI phase: increased odds of screening in DI pts. cf. EI phase. Increased prescription of AUD medication in EI pts. at DI phase.YYY
Mertens 2015 USA Alcohol [41]420,94655454Physician (PCP), Non-physician providers (NPP), Medical Aid (MA) arms. Training: PCP trained in all of SBIRT, MAs trained to ask screening question, NPPs trained to ask weekly drinking questions, AUD screener and BI and RT. All arms: Screening + automated prompts added to electronic health record; implementation support; audit + feedback.S, BI, RT3,4YHigher screening rates in NPP, MA and PCP cf. controls. Higher BI and referrals in PCP cf. other arms (No difference between NPP, MA and controls).NYY
Navarro 2012 AUS Alcohol [42]155,170na20Feedback letter to GP: prescription + community dependence rates, information on pharmacotherapies + behavioural interventions; recommendation to increase prescribing to reduce heavy alcohol consumption.RP4NIncreased acamprosate but decreased naltrexone prescribing cf. controls.NNN
Rose 2008 USA Alcohol [44]27,591na22NIAAA screening guidelines; instructions to develop/adapt screening template in electronic MRs; performance feedback and review; on-site visits with training, development of action plan; network meetings. Screening template (5As with AUDIT-C + question /diagnosis/recording prompts).S, BI, RT2,3,4YScreening, counselling odds higher in intervention cf. controls. Improvements over time greater in intervention arm. Reduced PB in pts. given brief counselling or referral but No significant reduction for intervention cf. controls.YYY
Saitz 2003 USA Alcohol [45]212411Clinical prompt: Results of CAGE assessment + recommendations attached to pt. record.BI, RT4NFaculty physicians in intervention arm more likely cf. controls to give advice, discuss associated problems. No significant difference in outcomes for residents in intervention cf. control arm. At 6-months, intervention arm pts. who saw residents had fewer drinks/drinking day but no between group differences.NNN
van Beurden 2012 NLD Alcohol [46]1502124823 components targeting: [1] professionals: training, guidelines, reminder cards [2]; organization: feedback report, facilitation of external specialist support, implementation support [3]; Pt-directed: letters, leaflets, self-help booklets, poster, personal feedback based on consumption.S, BI3,4,5YNo difference in improvement in screening or BI in intervention cf. control.?YN
Non randomized controlled studies
 Bradley 2002 USA Alcohol [47]68342Clinical prompt: pt-specific positive screening result at each visit.BI4NIntervention group more likely to discuss alcohol use cf. controls.NNN
 Hamilton 2014 GBR Alcohol [48]211,834na30Pay-for-performance scheme (QOF+) to extend alcohol screening; computer templates; in-practice training.S, BI, RT2YIncreased screening in eligible and Non-eligible group cf. baseline. Eligible pts. more likely to receive ASBI and full AUDIT than Not eligible.NNN
Harris 2017 USA Alcohol [49]295219933 components targeting [1] Local champion: training + support, monthly teleconferences, access to national champions, website, pt. dashboard [2]; Providers: training + support, website, access to local champions, pt. dashboard, reminder emails [3]; Pt education + activation: mailed materials.RP2,3,4,5YIncreased odds of filling a prescription during implementation in the three sites, however not significant at one site when stratified by site. No significant changes cf. matched controls.N?N
Khadjesari 2017 GBR Alcohol [50]261,424nanaPay-for-performance scheme for specific clinical areas.S1,2NIncrease in alcohol recording rate ratio over 13 years in case group cf. control group.NNN
Mason 1997 GBR Alcohol [51]1417na4Nurse-counsellor providing counselling services to practices + training to physicians.S, BI, RT3YIncrease in: recording of consumption and identification of problem drinkers (all intervention sites), identification of pts. drinking above recommended limits and advice (2 sites). No increases in referrals.NNN
McElwaine 2014 AUS Broad prev [52].198957017Local leadership engagement, electronic MR modification, training, implementation support, audit + feedback.S, BI, RT2,3,4YIncrease in odds of provision of Brief Advice from baseline to follow-up in intervention cf. controls. No changes in screening or referrals.Y?Y
O’Donnell 2016 GBR Alcohol [53]106,7009916Two pay-for-performance schemes: National (DES) - for each newly registered pt. screened; Local (LES) - for each new pt. over 16 positive for risky drinking + received BI.S, BI1,2NRates of short screening (FAST or AUDIT-C) or AUDIT lowest in non-incentivised and highest in DES. Rates of alcohol intervention lowest in non-incentivised and highest in DES. Significance Not reported.NNN
Onders 2014 USA Broad prev [54].23,000 visits/year Indigenous101 (cf national service)Electronic clinical reminders (CR) using PDSA: [1] data-driven ID of need [2]; Pilot test CR [3]; Expand to all providers [4]. audit +feedback [5]; Delegation of CR to other staff.S3,4YIncreased screening from 35 to 70% cf. IHS (smaller increase 40–48%) cf. other IHS.YYY
Ozer 2005 USA Broad prev [55].T0 = 226(i), 246(c); T1 = 551(i), 260(c); T2 = 940(i), 405(c) Adolescents764Clinician training; facilitated implementation of screening and charting forms tailored to local conditions. (Setting: paediatric PHC).S, BI3,4YIncreased screening and counselling post implementation both elements (intervention cf. controls). Increases associated with post training. No additional increases post tool implementation.N?N
Thomas 2014 SWE Broad prev [56].T1 = 888 T2 = 994T1 = 120 T2 = 1326Implementing screening for risky behaviour + BI and referrals to in-house multidisciplinary team; compulsory components: multidisciplinary teams + managers, meetings, in-house referral workflows.S, BI3,4YNo difference in alcohol consumption discussion rates in intervention cf. control at 3 years. Significantly higher alcohol discussion rates in control cf. intervention at 5 years.N?N
Wiggers 2017 AUS Broad prev [57].5369~ 140056Policy + leadership engagement; modifying information systems; training; audit and feedback; implementation support, information and resources.S, BI, RT2,3,4YIncreased alcohol consumption assessment and advice cf. control (stepped wedge). No increases in referrals.YYY
Wilson 1992 GBR Broad prev [58].44711610Increased consultation booking time from 6 to 10 min per pt.na3NIncreased recording of alcohol education and in pt-reported discussion about alcohol.NNN
Before/after and other designs
Healthy Habits
 Seale 2005a USA Alcohol [77]3387351Formation of lead committee (monitoring + recommendations); strategy development, modification of pt. record + workflows to include SBI; clinician training.S, BI, RT3,4YIncreased screening + intervention. Clinicians intervened more often when prompted with AUDITs. Periodic evaluation resulted in modifications, which resulted in progressive increases in screening.YYY
 Seale 2005b USA Alcohol [76]1052381Formation of leading committee (monitoring + recommendations); strategy development, modification of pt. record + workflows to include SBI; clinician training.S, BI4YNo significant differences in problem drinking (PD) identification before and after intervention. After training, greater increase in advice giving in residents cf. faculty.YYY
 Johnson 2013 USA Alcohol [69]288na1SBI workflow additions to protocols implemented above: Single Alcohol Screening Question (SASQ); checkbox in pt. record for BI; booster training.S, BI3,4YScreening rates using AUDIT-C plus SASQ exceeded 90% but no significant changes. Increased identification of UAU at 6 weeks and 6 months.???
VA program
Lapham 2012 USA Alcohol [70]6788nanaHealth system-wide incentives-linked performance measure (PM) + BI electronic clinical reminder (CR); freedom of adaptation but core PM components required; BI clinician training optional.BI2,3,4YRecording of advice increased continuously from baseline year, after PM announcement, PM implementation, and CR dissemination.NYN
Chavez 2016 USA Alcohol [62]225,912nanaAs above.BI2,3,4YIncreased pt-reported advice to pts. with moderate-severe alcohol misuse from baseline with plateau in two final years.NYN
 Si 2007 AUS Broad prev [78].360 Indigenousna12Single intervention cycle includes: initial systems audit + records audit; identify priorities + design improvement strategy; audit + feedback to monitor + identify new priorities.BI3YAlcohol counselling/advice increased significantly at Year 2 audit.YYY
 Gibson-Helm 2016 AUS Broad prev [65].2220 (Indigenous Pregnant women)na50As above.S, BI3,4YIncreased odds of screening and BI with each cycle. Evidence of a trend in increased number of CQI cycles and increase in BI.YYY
 Aalto 2003 Finland Alcohol [59]1449242Collaborative BI implementation development; training; implementation support; reminders in local and professional publications; distribution of AUDIT to all households.S, BI2,3,4,5YNo statistically significant differences.N?N
 Aspy 2008 USA Broad prev [60].600309Audits + feedback; training; practice facilitation/support; meetings between participating practice teams; Nurses, medical assistants, trained in screening and very brief interventions (VBI) e.g. referrals and handouts. Clinicians trained in BI.S, BI2,3,4YNo significant changes in alcohol screening or VBI or BI cf. baseline. Pts less likely to screen positive for UAU at end of study cf. baseline. Screening increased if alcohol was the target in first two cycles. Addition of more than two target behaviours appeared to negatively impact previous targets.YYY
Bobb 2017 USA Alcohol [61]53,133na33 strategies: [1] Enabling teams: recruitment + CQI training of site champions, development + implementation support, regular education on CQI, SBI + AUD treatment, information sharing between sites [2]. Support via electronic health record: screening, BI, AUD prompts [3]; Monitoring + feedback: PDSA; meetings.S, BI, RP, RT2,3,4YIncreased screening and assessment post-implementation cf. baseline. Effects sustained 1 year later + increased new AUD diagnoses. Increased treatment within 30- and 90-days post diagnosis (driven mainly by one site).YYY
Clifford 2013 AUS Alcohol [63]9322na4Training; treatment guidelines; electronic assessment tool; implementation supportS, BI3,4YIncreased records of screening. Increased records of BI overall but Not significant in individual ACCHS.NYN
Cowan 1994 USA Alcohol [64]910111Clinician training.Y
4NSignificant increase in recording of drinking histories.NNN
Gilkes 2017 AUS Broad prev [66].260835naClinical audit + feedback by medical students to GP supervisors.S4NIncreased record of alcohol consumption.NNY
Gowin 2012 POL Broad prev [67].1060106naRegional training program.S2,4NIncreased screening.NNN
Holtrop 2009 USA Broad prev [68].1965na20Record audits + practice assessment; choice of improvement plan based on 5As, priority risk behaviour or both; support in planning + implementation; audit + feedback at end of study.S, BI3,4YNo practice chose alcohol as target. However, increased alcohol screening but non-significant when adjusted for clustering.YY?
Lawner 1997 USA Alcohol [71]297151Training of faculty members to give performance feedback to residents with a feedback form.S4NIncreased record of alcohol consumption. Increased use of CAGE.NNN
Lustig 2001 USA Broad prev [72].532 (Adolescents)633Clinician training (Setting: paediatric PHC).S, BI4NIncreased screening.NNN
Marco-Garcia 1999 ESP Broad prev [73].(1500–2000/doctor, 42 doctors)843Formation of task force; collaborative program development; consensus on indicators + evaluation criteria; regular audit; action in response to audit.S3,4YIncreased recording of alcohol consumption.YYY
Olfson 1992 USA Alcohol [74]8841101Clinical prompt: addition of CAGE to health form completed by first-time pts. prior to first consult.S4NAlcohol problem detection (either problem drinking or abuse) increased cf. baseline.NNN
Seale 2015 USA Alcohol [75]1318na4Partial funding for coordinator; coordinator + clinician training; implementation committees; implementation guide + freedom to adapt to local setting; progress feedback.S, BI, RP, RT3,4YIncreased record of any screening or validated screening. Increased identification of risky users. Increased record of BI.?Y?
  1. aAuthor, year, country (as three-letter ISO 3166 country codes), focus and citation are given; bsignificant positive result for clinical action is indicated in bold; Y – Yes, N – No; S – screening; BI – brief intervention; RP – relapse prevention medicines; PT – psychosocial therapies; RT – referral to treatment; Strategy targets: Clin. Action – clinical action, Org. level – organisational level (1 = National, 2 = Health System, 3 = Practice, 4 = Clinician, 5 = Patient), M/faceted – multifaceted; cCQI elements: 1- Using ‘systematic data guided activities’ to identify problems and achieve improvement; 2 - ‘designing with local conditions in mind’ i.e. adapting and or designing strategies to fit the special characteristics of the local setting; 3 - using an ‘iterative development and testing process’; na – not available in article; (c) – control; (i) – intervention; pt. – patient; MR – medical record, eMR electronic medical record,.? – unclear; cf. – compared with