Treatment for Alcohol Dependence in Primary Care Compared to Outpatient Specialist Treatment—A Randomized Controlled Trial

Treatment for Alcohol Dependence in Primary Care Compared to Outpatient Specialist Treatment—A... Abstract Aim To investigate if treatment for alcohol dependence in primary care is as effective as specialist addiction care. Method Randomized controlled non-inferiority trial, between groups parallel design, not blinded. The non-inferiority limit was set to 50 grams of alcohol per week. About 288 adults fulfilling ICD-10 criteria for alcohol dependence were randomized to treatment in primary care (men n = 82, women n = 62) or specialist care (men n = 77, women n = 67). General practitioners at 12 primary care centers received 1-day training in a treatment manual for alcohol dependence. Primary outcome was change in weekly alcohol consumption at 6-months follow-up compared with baseline, as measured with timeline follow back. Secondary outcomes were heavy drinking days, severity of dependence, consequences of drinking, psychological health, quality of life, satisfaction with treatment and biomarkers. Results Intention-to-treat analysis (n = 228) was statistically inconclusive, and could not confirm non-inferiority for the primary outcome, since the high end of the confidence interval exceeded 50 grams (estimated mean weekly alcohol consumption was 30 grams higher in primary care compared with specialist care; 95% confidence interval −10.20; 69.72). However, treatment in specialist care was not significantly superior to primary care (P = 0.146). Subanalysis suggests that specialist care was superior to primary care only for patients with high severity of dependence. Conclusions Treatment for alcohol dependence in primary care is a promising approach, especially for individuals with low to moderate dependence. This may be a way to broaden the base of treatment for alcohol dependence, reducing the current treatment gap. INTRODUCTION Though treatment seeking has been found to increase the rates of recovery from alcohol dependence (Trim et al., 2013) fewer than one in five seek treatment (Rehm et al., 2015a). Alcohol dependence has one of the largest treatment gaps between the number of individuals affected and the number in treatment (Kohn et al., 2004). Treatment seekers typically have a more severe form of dependence, often in combination with co-morbid disorders and a more unstable social situation, e.g. homelessness or unemployment (Kohn et al., 2004; Storbjork and Room, 2008). In addition, individuals with moderate alcohol dependence compared to more severe forms of dependence tend to have less psychiatric co-morbidities, a more stable social situation and more rarely seek treatment. Several treatments have shown efficacy in reducing alcohol consumption for individuals with dependence. In the EU, estimates show that 12,000 lives could annually be saved by treating two out of five alcohol dependent individuals, rather than today's one in five (Rehm et al., 2013). Thus, there is a need for the treatment services to reach and treat a larger proportion of individuals with excessive alcohol consumption. An important barrier to seek treatment is the stigma surrounding alcohol dependence and treatment seeking (Wallhed Finn et al., 2014; Probst et al., 2015). One possible way of reducing the stigma is to offer treatment for alcohol dependence in primary care. Studies indicate that individuals with alcohol dependence are positive to seeking treatment in primary care (Lieberman et al., 2014; Wallhed Finn et al., 2014). In addition, a large proportion of individuals with alcohol dependence are already in primary care treatment for other conditions (Rehm et al., 2015a). Research on interventions for alcohol problems in primary care have mainly focused on Screening, Brief Interventions and Referral to Treatment, also known as SBIRT (O’Donnell et al., 2014). However, there is a lack of evidence that screening and brief interventions lead to a higher rate of treatment utilization in specialist care among problem drinkers, independent of problem severity (Glass et al., 2015). Most treatment studies in primary care have included individuals with hazardous or harmful consumption but alcohol dependent patients have been studied to a lesser extent. A British study found no difference in reduction of alcohol consumption for patients with alcohol dependence treated by general practitioners compared to specialist treatment (Drummond et al., 1990). However, the general practitioners received continuous support from the specialists during the course of treatments. There was also a significant cross over between the treatment arms. A stepped care intervention delivered via telephone has been shown to reduce alcohol consumption among at-risk drinkers in primary care (Bischof et al., 2008). However, no effects were found among individuals with alcohol dependence. Longitudinal care with Chronic Care Management in primary care, targeting individuals with severe dependence, has shown mixed results compared to treatment as usual in primary care (Willenbring and Olson, 1999; Saitz et al., 2013; Upshur et al., 2015). A recent study has shown that care based on similar principles increased treatment utilization and abstinence among individuals with opioid and alcohol dependence (Watkins et al., 2017). Two studies have investigated treatment with naltrexone combined with behavioural interventions in primary care compared to specialist care (O’Malley et al., 2003; Oslin et al., 2013). In both studies, the behavioural interventions were delivered by generalists without specialist training in treatment for alcohol dependence, and the caregivers were regularly supervised. O’Malley et al. (2003) found an advantage for specialist care on percentage of days abstinent, but no differences on secondary drinking outcomes. Oslin et al. (2013) found treatment in primary care more effective in reducing heavy drinking days (HDD) compared to specialist care. Moreover, retention to treatment was significantly higher in primary care. Three studies have evaluated the effects of pharmacological treatment in primary care delivered by generalists without specialist training. Kiritze-Topor et al. (2004) showed that adding acamprosate to treatment as usual, resulted in a longer duration of abstinence. However, a study from North Carolina and Wisconsin showed no significant effects of acamprosate compared to placebo, when delivered concomitant to five sessions of brief intervention (Berger et al., 2013). As-needed nalmefene, together with a minimal psychosocial intervention, was found to reduce the number of HDD compared to placebo (Karhuvaara et al., 2007). Thus, there is preliminary evidence for successful treatment of alcohol dependence by generalists in primary care. However, as time constraints are an important issue in primary care, one important challenge is to develop brief and effective treatment models that are feasible to implement. Also, in regular clinical practice, access to supervision from specialists is scarce, emphasizing the need of evaluating interventions delivered by generalists without supervision. As stand-alone interventions, brief interventions have been shown to reduce alcohol consumption among hazardous drinkers, but not individuals with alcohol dependence (Saitz, 2010). We have developed a manual for treatment of hazardous drinking and alcohol dependence in primary care, the 15-method. The manual starts with a brief intervention and, if needed, the patient proceeds to interventions that are more extensive with the possibility of receiving pharmacological and brief psychological treatment (Riddargatan 1, 2012). The model consists of three steps: (1) identification of problem drinking and brief advice (Alvarez-Bueno et al., 2015); (2) assessment, with a 30-min feedback session (Chick et al., 1988; Miller et al., 1988); and (3) four sessions based on CBT and motivational interviewing (Sobell and Sobell, 1993; Andreasson et al., 2002). Each session contains a theme to facilitate behaviour change: goal setting, self-monitoring of alcohol consumption, identifying risk situations and problem solving. These sessions can be combined with pharmacological treatment; acamprosate, disulfiram, nalmefene or naltrexone. The name, ‘the 15-method’ refers to the length of sessions, 15 min, and that the final two steps of the intervention target patients who score >15 points on the AUDIT (Babor et al., 2001). Shared decision-making guides the type of treatment, treatment goal and intensity of treatment provided. We have found no study of brief treatment, integrated with the choice of the full range of available pharmacological treatment delivered in primary care, adapted to the context and time constraints of primary care. Furthermore, to our knowledge there are no studies of treatment for alcohol dependence delivered by general practitioners without access to continuous support and supervision from specialists. Objectives The overall objective of the present study was to investigate if treatment for alcohol dependence in primary care was as effective as standard treatment delivered in specialist care. We will present results from a qualitative study, where practitioners and managers at the primary health care centres involved in this study were interviewed, in a separate publication. Hypothesis Treatment for alcohol dependence in primary care with the 15-method was non-inferior to specialist care. METHODS The study was approved by the regional ethics board in Stockholm, 2012-11-07, ref: 2012/1760-31/1. Trial design Randomized controlled non-inferiority trial, between groups parallel design 1:1 allocation ratio. Participants Participants were recruited via advertisement in newspapers, leaflets at the primary care study sites or direct question at a primary care consultation. After an initial telephone screening with a research nurse, eligible individuals, who preliminarily fulfilled the ICD-10 criteria for alcohol dependence, were scheduled for an assessment at the research unit. At the face-to-face assessment, a more detailed assessment of the ICD-10 criteria was conducted. Standardized protocols for both the telephone and the face-to-face assessments were used. The research nurse was trained by the third author in conducting the assessment protocol. The inclusion criteria were as follows: alcohol dependence according to ICD-10; ≥18 years old and living within Stockholm county. The exclusion criteria were as follows: need of continuous support from the social services (homeless, financial support); previous severe withdrawal symptoms (seizures, hallucinations, delirium); misuse or dependence of narcotics or prescription drugs (ICD-10); severe somatic or psychiatric condition; or not able to understand Swedish. Interventions Primary care The manual was adapted from the 15-method (Riddargatan 1, 2012), which is a model for treatment of hazardous drinking and alcohol dependence in primary care. For this study, where participants were mainly recruited by advertisement, and had already completed intake assessment as a part of study procedures, the first step of the manual was redundant. Shared decision-making between the patient and physician guided whether participants received step 3 and which type of treatment (psychological, pharmacological or a combination), and also the treatment goal; abstinence or controlled drinking. The treatment according to the manual consisted of a maximum of five sessions or 90 min over 24 weeks. Specialist treatment Participants randomized to specialist treatment, first received feedback of the baseline assessment, delivered by a physician. Participants were then offered pharmacological and/or psychological treatment, and shared decision-making also guided the choice of treatment. The pharmacological treatments were identical to the ones in primary care regarding substances and dosage, however, with 30 min per visit. The manual based psychological treatments were either motivational enhancement treatment (four sessions) (Sellman et al., 2001), guided self-change (four sessions) (Sobell and Sobell, 1993), relapse prevention (six to eight sessions) (Magill and Ray, 2009), behavioural self-control training (four to five sessions) (Walters, 2000) or 12-step treatment (12 sessions) (Project MATCH, 1997). Each session in the psychological treatments lasted 45 min. All staff in both condition indicated the grade of fidelity to the treatment manuals by self-report at the end of each session. Settings The interventions in primary care were delivered by the regular general practitioners, who received an eight hour training in the 15-method. The training consisted of lectures on the manual, combined with skills training including role-play of cases. In total, 29 general practitioners at 12 different primary care centres delivered the treatment. Specialist care was delivered by the clinical staff at Riddargatan 1, an alcohol treatment clinic within the Stockholm Centre for Dependence Disorders. The specialist care was delivered by physicians, nurses, social workers, psychotherapists, psychologists and a 12-step therapist, in total 16 different staff members all with at least several years’ experience in treating alcohol dependence. Outcomes The primary outcome was change in weekly alcohol consumption measured in grams of alcohol before inclusion compared to 6 months after start of treatment. The secondary outcomes were: HDD (women > 3/men > 4 standard units of 12 g of alcohol) per month; severity of dependence; consequences of drinking; symptoms of anxiety and depression; health related quality of life; satisfaction with treatment and biomarkers. An independent research nurse performed baseline and 6-month follow-up measures with the participants. Measurements Questionnaires Time Line Follow Back 30 days (Sobell and Sobell, 1996) was used to assess weekly alcohol consumption and HDD. Severity of alcohol dependence was assessed with Alcohol Use Disorders Identification Test (AUDIT) (Babor et al., 2001); the ICD-10 criteria for alcohol dependence (The World Health Organization, 1992) and Short Alcohol Dependence Data Questionnaire (SADD) (Raistrick and Davidson, 1983). As AUDIT and the ICD-10 criteria assess the last 12 months, these questionnaires were re-worded for the 6-month follow up to only cover the last 6 months. The Short Index of Problems (SIP) assessed consequences of drinking (Miller et al., 1995). Symptoms of anxiety and depression were assessed with Hospital Anxiety and Depression Scale (HADS) (Zigmond and Snaith, 1983). The EQ 5D-5L questionnaire measured health related quality of life (Herdman et al., 2011). Satisfaction with treatment was assessed with Client Satisfaction Questionnaire (CSQ) (Larsen et al., 1979). Readiness to change was assessed with a visual analogue scale (range: 0–100), in which importance and confidence to reach your goal [regarding alcohol] were rated. Participants indicated whether they completed the treatment they were randomized to. Biomarkers Blood was analysed for levels of carbohydrate-deficient transferrin (CDT); gamma-glutamyl transferase (GGT); aspartate amino transferase (AST) and alanine amino transferase (ALT). Non-inferiority limits On the pre-specified primary outcome (difference in weekly alcohol consumption at baseline compared to 6 months following start of treatment), the non-inferiority limit was set to 50 g of alcohol per week. This limit was derived from previous research in our group (Andreasson et al., 2002). In this study, socially well-adjusted heavy drinkers, with moderate levels of dependence severity, received either four sessions of guided self-change or one session of advice. A difference in alcohol consumption of 50 g, or approximately four standard drinks, between groups, was deemed clinically non-significant in this heavy drinking population. Sample size Sample size was calculated for the primary outcome. With a non-inferiority limit set at 50 g of alcohol per week, assuming a standard deviation of 140 g, based on Andreasson et al., 2002, and a difference between groups of 0, a total of 250 patients were necessary to accomplish 80% power to confirm non-inferiority at a 95% confidence level. Assuming a dropout rate of around 20%, the aim was to include 300 participants. Randomization Participants who gave informed consent and completed intake measures were randomized to treatment either at primary care or specialist care. The allocation sequence was generated by the first author, using a computer program (Excel 2010). The sequence was in blocks of eight, and stored with restricted access only by administrative staff not involved in the study. Participants randomized to primary care chose the primary care centre that was geographically most convenient for them. Blinding Neither participants, nor researchers, nor treatment staffs were blinded to the allocations. The study design as a non-inferiority study was known among participants, care providers and outcome assessors. Statistical methods The primary analysis was based on the intention-to-treat (ITT) data set, using data from all available cases as randomized, i.e. all participants who completed follow-up measures at 6 months regardless of protocol adherence. The analysis of the primary outcome was performed using an analysis of covariance (ANCOVA) with the consumption at 6 months as the dependent variable, group as a factor variable, adjusting for baseline consumption. The same statistical model was used for analyses of secondary outcomes. Factors related to the primary outcome was analysed with a regression model. Additional analyses used t-tests and chi-square tests. All analyses used two-sided tests and a significance level of P < 0.05. In-group effect sizes were calculated with Cohen's d. Two sensitivity analyses were performed on the primary outcome. In the first, the ITT data set was used with multiple imputation by predictive mean matching (Witkiewitz et al., 2014). The second analysis was based on a per protocol (PP) principle, where participants with at least one treatment visit, no concurrent treatment for alcohol dependence during the follow up period and who completed the follow-up measures were included. All statistical analyses were conducted with R, Version 3.1.0. RESULTS Participant flow The CONSORT flow chart is shown in Fig. 1. Fig. 1. View largeDownload slide Flow diagram of study enrolment, allocation, 6-month follow-up and analysis. Fig. 1. View largeDownload slide Flow diagram of study enrolment, allocation, 6-month follow-up and analysis. Participants were enroled between October 2013 and March 2015. Due to time constraint, data collection ceased when 288 of the targeted 300 participants had been included. One participant was excluded after randomization to primary care but before first visit due to high liver enzymes. The 6-month follow-up period lasted between May 2014 and January 2016. Table 1 shows demographic and clinical characteristics of each group. Table 1. Baseline characteristics for participants in primary care and specialist care (n = 288)   Primary care (n = 144)  Specialist care (n = 144)  P-value  Gender [n (%)]      0.636   Male  82 (57)  77 (53)     Female  62 (43)  67 (47)    Age [mean (SD) range]  56 (11) 23–77  54 (12) 25–79  0.094  Education [n (%)]      0.031   Not completed compulsory education  2 (1)  0 (0)     9 Years of education  13 (9)  4 (3)     12 Years of education  50 (35)  59 (41)     Higher education  78 (54)  76 (53)     Other  1 (1)  5 (3)    Source of income [n (%)]      1.000   Employment  105 (74)  104 (73)     Pension  31 (22)  31 (22)     Other  6 (4)  7 (5)    Civil status [n (%)]      0.223   Married/co-habiting  83 (58)  92 (64)     Live alone  58 (41)  48 (34)     Widowed  2 (1)  3 (2)    Use of nicotine (current) [n (%)]      0.737   Cigarettes  24 (17)  28 (19)     Snuff  24 (17)  20 (14)     Cigarettes and snuff  7 (5)  10 (7)    Use of illicit drugs (lifetime) [n (%)]  60 (42)  60 (42)  1.000  Problematic alcohol use (years) [mean (SD) range]  11.4 (10.5) 1–55  11.5 (10.7) 0.1–50  0.911  Treatment naive [% (n)]  104 (75)  113 (80)  0.337  Weekly alcohol consumption (g) [mean (SD) range]  343.62 (206.58) 0–1173.2  349.16 (216.45) 0–1680  0.824  Heavy drinking days per month [mean (SD) range]  12.23 (7.81) 0–30  11.92 (7.98) 0–30  0.737  ICD-10 criteria dependence [mean (SD) range]  4.22 (1.01) 2–6  4.39 (1.05) 3–6  0.169  AUDIT total score [mean (SD) range]  22.57 (5.75) 6–37 (n = 141)  23.49 (4.97) 10–38 (n = 138)  0.152  SADD total score [mean (SD) range]  14.83 (6.33) 3–40 (n = 140)  14.39 (5.23) 4–33 (n = 137)  0.535  SIP total score [mean (SD) range]  15.64 (7.48) 2–38 (n = 142)  16.21 (6.64) 3–34 (n = 143)  0.497  HADS scale anxiety [mean (SD) range]  7.36 (4.26) 0–19 (n = 143)  7.58 (4.45) 0–19 (n = 143)  0.674  HADS scale depression [mean (SD) range]  4.71 (3.52) 0–16 (n = 143)  4.84 (3.34) 0–15 (n = 143)  0.743  EQ 5D 5 L VAS [mean (SD) range]  63.11 (17.98) 10–95 (n = 142)  64.49 (18.12) 10–100 (n = 144)  0.518  CDT [mean (SD) range]  2.45 (2.13) 0.13–13.3 (n = 142)  2.28 (1.84) 0.7–13.9 (n = 142)  0.464  GT [mean (SD) range]  0.79 (1.33) 0.1–11.8 (n = 143)  0.86 (1.43) 0.1–9.4 (n = 143)  0.639  ASAT [mean (SD) range]  0.54 (0.35) 0.22–2.8 (n = 143)  0.52 (0.35) 0.22–2.72 (n = 143)  0.616  ALAT [mean (SD) range]  0.57 (0.39) 0.1–2.22 (n = 143)  0.54 (0.42) 0.11–2.73 (n = 143)  0.583  Alcohol consumption below the Swedish national guidelines for hazardous consumption (HDD = 0 and men ≤14 units/week; women ≤9 units/week (1 unit = 12 g alcohol) [n (%)])  6 (4.17)  4 (2.78)  0.520  Readiness to change (importance)  93.60 (9.67) 60–100 (n = 138)  91.88 (11.56) 50–100 (n = 141)  0.179  Readiness to change (confidence)  77.91 (19.99) 0–100 (n = 137)  76.68 (21.03) 5–100 (n = 141)  0.620    Primary care (n = 144)  Specialist care (n = 144)  P-value  Gender [n (%)]      0.636   Male  82 (57)  77 (53)     Female  62 (43)  67 (47)    Age [mean (SD) range]  56 (11) 23–77  54 (12) 25–79  0.094  Education [n (%)]      0.031   Not completed compulsory education  2 (1)  0 (0)     9 Years of education  13 (9)  4 (3)     12 Years of education  50 (35)  59 (41)     Higher education  78 (54)  76 (53)     Other  1 (1)  5 (3)    Source of income [n (%)]      1.000   Employment  105 (74)  104 (73)     Pension  31 (22)  31 (22)     Other  6 (4)  7 (5)    Civil status [n (%)]      0.223   Married/co-habiting  83 (58)  92 (64)     Live alone  58 (41)  48 (34)     Widowed  2 (1)  3 (2)    Use of nicotine (current) [n (%)]      0.737   Cigarettes  24 (17)  28 (19)     Snuff  24 (17)  20 (14)     Cigarettes and snuff  7 (5)  10 (7)    Use of illicit drugs (lifetime) [n (%)]  60 (42)  60 (42)  1.000  Problematic alcohol use (years) [mean (SD) range]  11.4 (10.5) 1–55  11.5 (10.7) 0.1–50  0.911  Treatment naive [% (n)]  104 (75)  113 (80)  0.337  Weekly alcohol consumption (g) [mean (SD) range]  343.62 (206.58) 0–1173.2  349.16 (216.45) 0–1680  0.824  Heavy drinking days per month [mean (SD) range]  12.23 (7.81) 0–30  11.92 (7.98) 0–30  0.737  ICD-10 criteria dependence [mean (SD) range]  4.22 (1.01) 2–6  4.39 (1.05) 3–6  0.169  AUDIT total score [mean (SD) range]  22.57 (5.75) 6–37 (n = 141)  23.49 (4.97) 10–38 (n = 138)  0.152  SADD total score [mean (SD) range]  14.83 (6.33) 3–40 (n = 140)  14.39 (5.23) 4–33 (n = 137)  0.535  SIP total score [mean (SD) range]  15.64 (7.48) 2–38 (n = 142)  16.21 (6.64) 3–34 (n = 143)  0.497  HADS scale anxiety [mean (SD) range]  7.36 (4.26) 0–19 (n = 143)  7.58 (4.45) 0–19 (n = 143)  0.674  HADS scale depression [mean (SD) range]  4.71 (3.52) 0–16 (n = 143)  4.84 (3.34) 0–15 (n = 143)  0.743  EQ 5D 5 L VAS [mean (SD) range]  63.11 (17.98) 10–95 (n = 142)  64.49 (18.12) 10–100 (n = 144)  0.518  CDT [mean (SD) range]  2.45 (2.13) 0.13–13.3 (n = 142)  2.28 (1.84) 0.7–13.9 (n = 142)  0.464  GT [mean (SD) range]  0.79 (1.33) 0.1–11.8 (n = 143)  0.86 (1.43) 0.1–9.4 (n = 143)  0.639  ASAT [mean (SD) range]  0.54 (0.35) 0.22–2.8 (n = 143)  0.52 (0.35) 0.22–2.72 (n = 143)  0.616  ALAT [mean (SD) range]  0.57 (0.39) 0.1–2.22 (n = 143)  0.54 (0.42) 0.11–2.73 (n = 143)  0.583  Alcohol consumption below the Swedish national guidelines for hazardous consumption (HDD = 0 and men ≤14 units/week; women ≤9 units/week (1 unit = 12 g alcohol) [n (%)])  6 (4.17)  4 (2.78)  0.520  Readiness to change (importance)  93.60 (9.67) 60–100 (n = 138)  91.88 (11.56) 50–100 (n = 141)  0.179  Readiness to change (confidence)  77.91 (19.99) 0–100 (n = 137)  76.68 (21.03) 5–100 (n = 141)  0.620  Table 1. Baseline characteristics for participants in primary care and specialist care (n = 288)   Primary care (n = 144)  Specialist care (n = 144)  P-value  Gender [n (%)]      0.636   Male  82 (57)  77 (53)     Female  62 (43)  67 (47)    Age [mean (SD) range]  56 (11) 23–77  54 (12) 25–79  0.094  Education [n (%)]      0.031   Not completed compulsory education  2 (1)  0 (0)     9 Years of education  13 (9)  4 (3)     12 Years of education  50 (35)  59 (41)     Higher education  78 (54)  76 (53)     Other  1 (1)  5 (3)    Source of income [n (%)]      1.000   Employment  105 (74)  104 (73)     Pension  31 (22)  31 (22)     Other  6 (4)  7 (5)    Civil status [n (%)]      0.223   Married/co-habiting  83 (58)  92 (64)     Live alone  58 (41)  48 (34)     Widowed  2 (1)  3 (2)    Use of nicotine (current) [n (%)]      0.737   Cigarettes  24 (17)  28 (19)     Snuff  24 (17)  20 (14)     Cigarettes and snuff  7 (5)  10 (7)    Use of illicit drugs (lifetime) [n (%)]  60 (42)  60 (42)  1.000  Problematic alcohol use (years) [mean (SD) range]  11.4 (10.5) 1–55  11.5 (10.7) 0.1–50  0.911  Treatment naive [% (n)]  104 (75)  113 (80)  0.337  Weekly alcohol consumption (g) [mean (SD) range]  343.62 (206.58) 0–1173.2  349.16 (216.45) 0–1680  0.824  Heavy drinking days per month [mean (SD) range]  12.23 (7.81) 0–30  11.92 (7.98) 0–30  0.737  ICD-10 criteria dependence [mean (SD) range]  4.22 (1.01) 2–6  4.39 (1.05) 3–6  0.169  AUDIT total score [mean (SD) range]  22.57 (5.75) 6–37 (n = 141)  23.49 (4.97) 10–38 (n = 138)  0.152  SADD total score [mean (SD) range]  14.83 (6.33) 3–40 (n = 140)  14.39 (5.23) 4–33 (n = 137)  0.535  SIP total score [mean (SD) range]  15.64 (7.48) 2–38 (n = 142)  16.21 (6.64) 3–34 (n = 143)  0.497  HADS scale anxiety [mean (SD) range]  7.36 (4.26) 0–19 (n = 143)  7.58 (4.45) 0–19 (n = 143)  0.674  HADS scale depression [mean (SD) range]  4.71 (3.52) 0–16 (n = 143)  4.84 (3.34) 0–15 (n = 143)  0.743  EQ 5D 5 L VAS [mean (SD) range]  63.11 (17.98) 10–95 (n = 142)  64.49 (18.12) 10–100 (n = 144)  0.518  CDT [mean (SD) range]  2.45 (2.13) 0.13–13.3 (n = 142)  2.28 (1.84) 0.7–13.9 (n = 142)  0.464  GT [mean (SD) range]  0.79 (1.33) 0.1–11.8 (n = 143)  0.86 (1.43) 0.1–9.4 (n = 143)  0.639  ASAT [mean (SD) range]  0.54 (0.35) 0.22–2.8 (n = 143)  0.52 (0.35) 0.22–2.72 (n = 143)  0.616  ALAT [mean (SD) range]  0.57 (0.39) 0.1–2.22 (n = 143)  0.54 (0.42) 0.11–2.73 (n = 143)  0.583  Alcohol consumption below the Swedish national guidelines for hazardous consumption (HDD = 0 and men ≤14 units/week; women ≤9 units/week (1 unit = 12 g alcohol) [n (%)])  6 (4.17)  4 (2.78)  0.520  Readiness to change (importance)  93.60 (9.67) 60–100 (n = 138)  91.88 (11.56) 50–100 (n = 141)  0.179  Readiness to change (confidence)  77.91 (19.99) 0–100 (n = 137)  76.68 (21.03) 5–100 (n = 141)  0.620    Primary care (n = 144)  Specialist care (n = 144)  P-value  Gender [n (%)]      0.636   Male  82 (57)  77 (53)     Female  62 (43)  67 (47)    Age [mean (SD) range]  56 (11) 23–77  54 (12) 25–79  0.094  Education [n (%)]      0.031   Not completed compulsory education  2 (1)  0 (0)     9 Years of education  13 (9)  4 (3)     12 Years of education  50 (35)  59 (41)     Higher education  78 (54)  76 (53)     Other  1 (1)  5 (3)    Source of income [n (%)]      1.000   Employment  105 (74)  104 (73)     Pension  31 (22)  31 (22)     Other  6 (4)  7 (5)    Civil status [n (%)]      0.223   Married/co-habiting  83 (58)  92 (64)     Live alone  58 (41)  48 (34)     Widowed  2 (1)  3 (2)    Use of nicotine (current) [n (%)]      0.737   Cigarettes  24 (17)  28 (19)     Snuff  24 (17)  20 (14)     Cigarettes and snuff  7 (5)  10 (7)    Use of illicit drugs (lifetime) [n (%)]  60 (42)  60 (42)  1.000  Problematic alcohol use (years) [mean (SD) range]  11.4 (10.5) 1–55  11.5 (10.7) 0.1–50  0.911  Treatment naive [% (n)]  104 (75)  113 (80)  0.337  Weekly alcohol consumption (g) [mean (SD) range]  343.62 (206.58) 0–1173.2  349.16 (216.45) 0–1680  0.824  Heavy drinking days per month [mean (SD) range]  12.23 (7.81) 0–30  11.92 (7.98) 0–30  0.737  ICD-10 criteria dependence [mean (SD) range]  4.22 (1.01) 2–6  4.39 (1.05) 3–6  0.169  AUDIT total score [mean (SD) range]  22.57 (5.75) 6–37 (n = 141)  23.49 (4.97) 10–38 (n = 138)  0.152  SADD total score [mean (SD) range]  14.83 (6.33) 3–40 (n = 140)  14.39 (5.23) 4–33 (n = 137)  0.535  SIP total score [mean (SD) range]  15.64 (7.48) 2–38 (n = 142)  16.21 (6.64) 3–34 (n = 143)  0.497  HADS scale anxiety [mean (SD) range]  7.36 (4.26) 0–19 (n = 143)  7.58 (4.45) 0–19 (n = 143)  0.674  HADS scale depression [mean (SD) range]  4.71 (3.52) 0–16 (n = 143)  4.84 (3.34) 0–15 (n = 143)  0.743  EQ 5D 5 L VAS [mean (SD) range]  63.11 (17.98) 10–95 (n = 142)  64.49 (18.12) 10–100 (n = 144)  0.518  CDT [mean (SD) range]  2.45 (2.13) 0.13–13.3 (n = 142)  2.28 (1.84) 0.7–13.9 (n = 142)  0.464  GT [mean (SD) range]  0.79 (1.33) 0.1–11.8 (n = 143)  0.86 (1.43) 0.1–9.4 (n = 143)  0.639  ASAT [mean (SD) range]  0.54 (0.35) 0.22–2.8 (n = 143)  0.52 (0.35) 0.22–2.72 (n = 143)  0.616  ALAT [mean (SD) range]  0.57 (0.39) 0.1–2.22 (n = 143)  0.54 (0.42) 0.11–2.73 (n = 143)  0.583  Alcohol consumption below the Swedish national guidelines for hazardous consumption (HDD = 0 and men ≤14 units/week; women ≤9 units/week (1 unit = 12 g alcohol) [n (%)])  6 (4.17)  4 (2.78)  0.520  Readiness to change (importance)  93.60 (9.67) 60–100 (n = 138)  91.88 (11.56) 50–100 (n = 141)  0.179  Readiness to change (confidence)  77.91 (19.99) 0–100 (n = 137)  76.68 (21.03) 5–100 (n = 141)  0.620  Length and type of treatment Participants in primary care (n = 144) had on average fewer number of visits compared to specialist care (n = 144), 2.9 (SD = 1.5; range: 0–6) vs 4.7 (SD = 2.8; range: 0–14), (t = 6.913, P < 0.001). On average, participants in primary care (n = 141; missing data n = 3) also received shorter treatment in minutes; 74 (SD = 46; range: 0–195), compared to 187 (SD = 129; range: 0–685) in specialist care (n = 144), (t = 9.88, P = 0.000). In both treatment arms, a combination of pharmacological and psychological treatment was most common. In total, 76 participants in primary care and 62 participants in specialist care received this combination. The number of participants receiving only pharmacological treatment in primary care and in specialist care were 17 and 25, respectively. In general, there were no differences in which type of treatment patients received in the two treatment arms with two exceptions. Firstly, a larger number of patients in specialist care compared to primary care received only psychological treatment, (χ2 = 4.5, df = 1, P= 0.034). Secondly, a larger number of patients in specialist care compared to primary care received Antabuse (n = 28 respectively n = 12, χ2 = 9.4, df = 1, P = 0.002). Fidelity In primary care, the general practitioners reported they fully followed the manual in 89.6 % (n = 120) of the first visits, partially in 6.7% (n = 9) and for 3.7 % (n = 5) data were missing. For the psychological treatment in primary care, 93.6% (n = 192) of the visits were rated as fully followed the manual, and 6.3% (n = 13) as partially. Primary outcome The ITT analysis, with available cases, showed that patients treated in primary care (n = 109) reduced their average weekly alcohol consumption from 367 to 224 g at 6 months follow up. The corresponding decrease in specialist care (n = 119) was from 343 to 182 g. Controlling for baseline consumption, the results on the pre-specified non-inferiority limit of 50 g for weekly alcohol consumption at 6-month follow up, is statistically inconclusive. This is due to the high end of the confidence interval exceeds 50 g (estimated mean weekly alcohol consumption was 30 g higher in primary care compared to specialist care, 95% CI: −10.20; 69.72). However, treatment in specialist care was not significantly superior to primary care (P = 0.146). Effect sizes For the primary outcome, the effect size based on available cases, in primary care was 0.66 and for specialist care 0.74. Sensitivity analysis Sensitivity analyses were performed in order to test the robustness of the results. The first analysis, ITT with multiple imputation by predictive mean matching, showed that patients treated in primary care (n = 144) had an estimated mean weekly alcohol consumption that was 27 g higher compared to patients treated in specialist care (n = 144) (95% CI: −10.90; 65.69, P = 0.159). The PP analysis showed that patients treated in primary care (n = 103) had an estimated mean weekly consumption that was 32 g higher compared to patients treated specialist care (n = 115) (95% CI: −7.21; 70.85), P = 0.110). The results were consistent with the ITT analysis, and are statistically inconclusive, since the confidence intervals exceed 50 g. Primary and secondary outcomes Table 2 shows the pre-specified primary and secondary outcome measures at baseline and at 6 months follow-up for available cases. Table 2. Outcome measures at baseline and 6 months follow up for primary care and specialist treatment. ITT data set with available cases Variable [mean (SD) range] (n = analysed)  Primary care, baseline  Primary care, 6 months follow up  Specialist care, baseline  Specialist care, 6 months follow up  Primary care, difference  Specialist care, difference  P-value  Weekly alcohol consumption (g)  367.4 (215.8) 47.6–1173.2 (n = 109)  224.2 (224.5) 0–1268.4 (n = 109)  343.3 (217.0) 0–1680 (n = 119)  181.9 (150.7) 0–772.8 (n = 119)  −143.3 (202.6) −772.8 to 680.4 (n = 109)  −161.3 (168.5) −907.2 to 165.2 (n = 119)  0.464  Heavy drinking days (per month)  13.1 (8.0) 1–30 (n = 109)  7.9 (8.9) 0–30 (n = 109)  11.9 (8.0) 0–30 (n = 119)  6.8 (7.8) 0–30 (n = 119)  −5.2 (8.9) −30 to 19 (n = 109)  −5.1 (7.8) −30 to 16 (n = 119)  0.954  ICD-10 criteria dependence  4.2 (1.0) 3–6 (n = 109)  2.1 (1.7) 0–6 (n = 105)  4.3 (1.0) 3–6 (n = 119)  2.2 (1.7) 0–6 (n = 118)  −2.1 (1.8) −6 to 2 (n = 105)  −2.1 (1.8) −6 to 3 (n = 118)  0.900  AUDIT total score  22.4 (5.8) 6–37 (n = 106)  14.1 (7.0) 0–32 (n = 100)  23.2 (4.9) 10–33 (n = 115)  13.5 (6.5) 0–35 (n = 111)  −8.4 (5.9) −30 to 3 (n = 99)  −9.3 (7.6) −30 to 7 (n = 107)  0.325  SADD total score  14.4 (6.3) 3–35 (n = 105)  9.0 (5.7) 0–34 (n = 100)  14.2 (5.3) 4–33 (n = 115)  8.0 (4.3) 0–23 (n = 114)  −5.5 (5.2) −20 to 3 (n = 98)  −6.2 (5.8) −23 to 5 (n = 110)  0.392  SIP total score  14.9 (7.0) 2–32 (n = 108)  8.0 (5.8) 0–29 (n = 103)  16.0 (6.5) 3–33 (n = 118)  7.8 (5.6) 0–22 (n = 115)  −6.8 (6.4) −26 to 7 (n = 103)  −7.9 (7.1) −30 to 7 (n = 114)  0.234  HADS scale anxiety  6.9 (4.1) 0–19 (n = 108)  5.7 (3.9) 0–19 (n = 104)  7.3 (4.4) 0–19 (n = 118)  5.6 (3.5) 0–17 (n = 114)  −1.1 (3.3) −9 to 15 (n = 103)  −1.4 (3.8) −14 to 7 (n = 113)  0.535  HADS scale depression  4.3 (3.4) 0–16 (n = 108)  3.1 (2.6) 0–10 (n = 105)  4.6 (3.2) 0–15 (n = 118)  3.1 (2.9) 0–15 (n = 116)  −1.3 (2.8) −9 to 8 (n = 104)  −1.5 (3.4) −12 to 7 (n = 115)  0.629  EQ 5D 5 L VAS  63.2 (18.8) 10–95 (n = 107)  73.7 (16.9) 20–100 (n = 105)  64.9 (18.4) 10–100 (n = 119)  75.5 (15.0) 30–100 (n = 116)  10.4 (19.5) −70 to 55 (n = 103)  10.5 (16.1) −35 to 65 (n = 116)  0.990  CDT  2.5 (2.1) 0.6–10.8 (n = 108)  2.4 (2.0) 0.6–12.5 (n = 99)  2.3 (2.0) 0.9–13.9 (n = 117)  1.9 (1.7) 0.6–13.6 (n = 107)  −0.2 (1.6) −6.9 to 4.6 (n = 98)  −0.28 (1.00) −6.2 to 2.4 (n = 105)  0.668  GT  0.8 (1.4) 0.1–11.8 (n = 109)  0.6 (0.9) 0.1–5.6 (n = 99)  0.8 (1.3) 0.1–8.6 (n = 118)  0.6 (1.0) 0.1–7.4 (n = 107)  −0.2 (1.4) −11.3 to 2.7 (n = 99)  −0.2 (0.6) −2.8 to 1.1 (n = 106)  0.710  ASAT  0.5 (0.3) 0.2–1.6 (n = 109)  0.5 (0.4) 0.2–3.6 (n = 98)  0.5 (0.3) 0.2–2.1 (n = 118)  0.5 (0.3) 0.2–1.7 (n = 107)  −0.0 (0.4) −1.0 to 3.1 (n = 98)  −0.1 (0.3) −1.7 to 0.8 (n = 106)  0.808  ALAT  0.6 (0.4) 0.1–2.2 (n = 109)  0.5 (0.4) 0.1–2.7 (n = 99)  0.5 (0.4) 0.1–2.8 (n = 118)  0.5 (0.4) 0.1–2.5 (n = 107)  −0.0 (0.4) −1.7 to 2.3 (n = 99)  0.0 (0.3) −1.9 to 1.6 (n = 106)  0.569  CSQ total score    21.6 (6.0) 9–32 (n = 95)    26.1 (4.8) 12–32 (n = 109)      <0.001*  Proportion below the Swedish national guidelines for hazardous consumptiona [n (%)]  6 (4.2) (n = 109)  19 (17.4) (n = 109)  4 (2.8) (n = 119)  23 (19.3) (n = 119)  −13 (−13.2) (n = 109)  −19 (−16.5)(n = 119)  0.843 df(1)  Variable [mean (SD) range] (n = analysed)  Primary care, baseline  Primary care, 6 months follow up  Specialist care, baseline  Specialist care, 6 months follow up  Primary care, difference  Specialist care, difference  P-value  Weekly alcohol consumption (g)  367.4 (215.8) 47.6–1173.2 (n = 109)  224.2 (224.5) 0–1268.4 (n = 109)  343.3 (217.0) 0–1680 (n = 119)  181.9 (150.7) 0–772.8 (n = 119)  −143.3 (202.6) −772.8 to 680.4 (n = 109)  −161.3 (168.5) −907.2 to 165.2 (n = 119)  0.464  Heavy drinking days (per month)  13.1 (8.0) 1–30 (n = 109)  7.9 (8.9) 0–30 (n = 109)  11.9 (8.0) 0–30 (n = 119)  6.8 (7.8) 0–30 (n = 119)  −5.2 (8.9) −30 to 19 (n = 109)  −5.1 (7.8) −30 to 16 (n = 119)  0.954  ICD-10 criteria dependence  4.2 (1.0) 3–6 (n = 109)  2.1 (1.7) 0–6 (n = 105)  4.3 (1.0) 3–6 (n = 119)  2.2 (1.7) 0–6 (n = 118)  −2.1 (1.8) −6 to 2 (n = 105)  −2.1 (1.8) −6 to 3 (n = 118)  0.900  AUDIT total score  22.4 (5.8) 6–37 (n = 106)  14.1 (7.0) 0–32 (n = 100)  23.2 (4.9) 10–33 (n = 115)  13.5 (6.5) 0–35 (n = 111)  −8.4 (5.9) −30 to 3 (n = 99)  −9.3 (7.6) −30 to 7 (n = 107)  0.325  SADD total score  14.4 (6.3) 3–35 (n = 105)  9.0 (5.7) 0–34 (n = 100)  14.2 (5.3) 4–33 (n = 115)  8.0 (4.3) 0–23 (n = 114)  −5.5 (5.2) −20 to 3 (n = 98)  −6.2 (5.8) −23 to 5 (n = 110)  0.392  SIP total score  14.9 (7.0) 2–32 (n = 108)  8.0 (5.8) 0–29 (n = 103)  16.0 (6.5) 3–33 (n = 118)  7.8 (5.6) 0–22 (n = 115)  −6.8 (6.4) −26 to 7 (n = 103)  −7.9 (7.1) −30 to 7 (n = 114)  0.234  HADS scale anxiety  6.9 (4.1) 0–19 (n = 108)  5.7 (3.9) 0–19 (n = 104)  7.3 (4.4) 0–19 (n = 118)  5.6 (3.5) 0–17 (n = 114)  −1.1 (3.3) −9 to 15 (n = 103)  −1.4 (3.8) −14 to 7 (n = 113)  0.535  HADS scale depression  4.3 (3.4) 0–16 (n = 108)  3.1 (2.6) 0–10 (n = 105)  4.6 (3.2) 0–15 (n = 118)  3.1 (2.9) 0–15 (n = 116)  −1.3 (2.8) −9 to 8 (n = 104)  −1.5 (3.4) −12 to 7 (n = 115)  0.629  EQ 5D 5 L VAS  63.2 (18.8) 10–95 (n = 107)  73.7 (16.9) 20–100 (n = 105)  64.9 (18.4) 10–100 (n = 119)  75.5 (15.0) 30–100 (n = 116)  10.4 (19.5) −70 to 55 (n = 103)  10.5 (16.1) −35 to 65 (n = 116)  0.990  CDT  2.5 (2.1) 0.6–10.8 (n = 108)  2.4 (2.0) 0.6–12.5 (n = 99)  2.3 (2.0) 0.9–13.9 (n = 117)  1.9 (1.7) 0.6–13.6 (n = 107)  −0.2 (1.6) −6.9 to 4.6 (n = 98)  −0.28 (1.00) −6.2 to 2.4 (n = 105)  0.668  GT  0.8 (1.4) 0.1–11.8 (n = 109)  0.6 (0.9) 0.1–5.6 (n = 99)  0.8 (1.3) 0.1–8.6 (n = 118)  0.6 (1.0) 0.1–7.4 (n = 107)  −0.2 (1.4) −11.3 to 2.7 (n = 99)  −0.2 (0.6) −2.8 to 1.1 (n = 106)  0.710  ASAT  0.5 (0.3) 0.2–1.6 (n = 109)  0.5 (0.4) 0.2–3.6 (n = 98)  0.5 (0.3) 0.2–2.1 (n = 118)  0.5 (0.3) 0.2–1.7 (n = 107)  −0.0 (0.4) −1.0 to 3.1 (n = 98)  −0.1 (0.3) −1.7 to 0.8 (n = 106)  0.808  ALAT  0.6 (0.4) 0.1–2.2 (n = 109)  0.5 (0.4) 0.1–2.7 (n = 99)  0.5 (0.4) 0.1–2.8 (n = 118)  0.5 (0.4) 0.1–2.5 (n = 107)  −0.0 (0.4) −1.7 to 2.3 (n = 99)  0.0 (0.3) −1.9 to 1.6 (n = 106)  0.569  CSQ total score    21.6 (6.0) 9–32 (n = 95)    26.1 (4.8) 12–32 (n = 109)      <0.001*  Proportion below the Swedish national guidelines for hazardous consumptiona [n (%)]  6 (4.2) (n = 109)  19 (17.4) (n = 109)  4 (2.8) (n = 119)  23 (19.3) (n = 119)  −13 (−13.2) (n = 109)  −19 (−16.5)(n = 119)  0.843 df(1)  aHDD = 0 and men ≤14 units/week; women ≤9 units/week (1 unit = 12 g alcohol. Table 2. Outcome measures at baseline and 6 months follow up for primary care and specialist treatment. ITT data set with available cases Variable [mean (SD) range] (n = analysed)  Primary care, baseline  Primary care, 6 months follow up  Specialist care, baseline  Specialist care, 6 months follow up  Primary care, difference  Specialist care, difference  P-value  Weekly alcohol consumption (g)  367.4 (215.8) 47.6–1173.2 (n = 109)  224.2 (224.5) 0–1268.4 (n = 109)  343.3 (217.0) 0–1680 (n = 119)  181.9 (150.7) 0–772.8 (n = 119)  −143.3 (202.6) −772.8 to 680.4 (n = 109)  −161.3 (168.5) −907.2 to 165.2 (n = 119)  0.464  Heavy drinking days (per month)  13.1 (8.0) 1–30 (n = 109)  7.9 (8.9) 0–30 (n = 109)  11.9 (8.0) 0–30 (n = 119)  6.8 (7.8) 0–30 (n = 119)  −5.2 (8.9) −30 to 19 (n = 109)  −5.1 (7.8) −30 to 16 (n = 119)  0.954  ICD-10 criteria dependence  4.2 (1.0) 3–6 (n = 109)  2.1 (1.7) 0–6 (n = 105)  4.3 (1.0) 3–6 (n = 119)  2.2 (1.7) 0–6 (n = 118)  −2.1 (1.8) −6 to 2 (n = 105)  −2.1 (1.8) −6 to 3 (n = 118)  0.900  AUDIT total score  22.4 (5.8) 6–37 (n = 106)  14.1 (7.0) 0–32 (n = 100)  23.2 (4.9) 10–33 (n = 115)  13.5 (6.5) 0–35 (n = 111)  −8.4 (5.9) −30 to 3 (n = 99)  −9.3 (7.6) −30 to 7 (n = 107)  0.325  SADD total score  14.4 (6.3) 3–35 (n = 105)  9.0 (5.7) 0–34 (n = 100)  14.2 (5.3) 4–33 (n = 115)  8.0 (4.3) 0–23 (n = 114)  −5.5 (5.2) −20 to 3 (n = 98)  −6.2 (5.8) −23 to 5 (n = 110)  0.392  SIP total score  14.9 (7.0) 2–32 (n = 108)  8.0 (5.8) 0–29 (n = 103)  16.0 (6.5) 3–33 (n = 118)  7.8 (5.6) 0–22 (n = 115)  −6.8 (6.4) −26 to 7 (n = 103)  −7.9 (7.1) −30 to 7 (n = 114)  0.234  HADS scale anxiety  6.9 (4.1) 0–19 (n = 108)  5.7 (3.9) 0–19 (n = 104)  7.3 (4.4) 0–19 (n = 118)  5.6 (3.5) 0–17 (n = 114)  −1.1 (3.3) −9 to 15 (n = 103)  −1.4 (3.8) −14 to 7 (n = 113)  0.535  HADS scale depression  4.3 (3.4) 0–16 (n = 108)  3.1 (2.6) 0–10 (n = 105)  4.6 (3.2) 0–15 (n = 118)  3.1 (2.9) 0–15 (n = 116)  −1.3 (2.8) −9 to 8 (n = 104)  −1.5 (3.4) −12 to 7 (n = 115)  0.629  EQ 5D 5 L VAS  63.2 (18.8) 10–95 (n = 107)  73.7 (16.9) 20–100 (n = 105)  64.9 (18.4) 10–100 (n = 119)  75.5 (15.0) 30–100 (n = 116)  10.4 (19.5) −70 to 55 (n = 103)  10.5 (16.1) −35 to 65 (n = 116)  0.990  CDT  2.5 (2.1) 0.6–10.8 (n = 108)  2.4 (2.0) 0.6–12.5 (n = 99)  2.3 (2.0) 0.9–13.9 (n = 117)  1.9 (1.7) 0.6–13.6 (n = 107)  −0.2 (1.6) −6.9 to 4.6 (n = 98)  −0.28 (1.00) −6.2 to 2.4 (n = 105)  0.668  GT  0.8 (1.4) 0.1–11.8 (n = 109)  0.6 (0.9) 0.1–5.6 (n = 99)  0.8 (1.3) 0.1–8.6 (n = 118)  0.6 (1.0) 0.1–7.4 (n = 107)  −0.2 (1.4) −11.3 to 2.7 (n = 99)  −0.2 (0.6) −2.8 to 1.1 (n = 106)  0.710  ASAT  0.5 (0.3) 0.2–1.6 (n = 109)  0.5 (0.4) 0.2–3.6 (n = 98)  0.5 (0.3) 0.2–2.1 (n = 118)  0.5 (0.3) 0.2–1.7 (n = 107)  −0.0 (0.4) −1.0 to 3.1 (n = 98)  −0.1 (0.3) −1.7 to 0.8 (n = 106)  0.808  ALAT  0.6 (0.4) 0.1–2.2 (n = 109)  0.5 (0.4) 0.1–2.7 (n = 99)  0.5 (0.4) 0.1–2.8 (n = 118)  0.5 (0.4) 0.1–2.5 (n = 107)  −0.0 (0.4) −1.7 to 2.3 (n = 99)  0.0 (0.3) −1.9 to 1.6 (n = 106)  0.569  CSQ total score    21.6 (6.0) 9–32 (n = 95)    26.1 (4.8) 12–32 (n = 109)      <0.001*  Proportion below the Swedish national guidelines for hazardous consumptiona [n (%)]  6 (4.2) (n = 109)  19 (17.4) (n = 109)  4 (2.8) (n = 119)  23 (19.3) (n = 119)  −13 (−13.2) (n = 109)  −19 (−16.5)(n = 119)  0.843 df(1)  Variable [mean (SD) range] (n = analysed)  Primary care, baseline  Primary care, 6 months follow up  Specialist care, baseline  Specialist care, 6 months follow up  Primary care, difference  Specialist care, difference  P-value  Weekly alcohol consumption (g)  367.4 (215.8) 47.6–1173.2 (n = 109)  224.2 (224.5) 0–1268.4 (n = 109)  343.3 (217.0) 0–1680 (n = 119)  181.9 (150.7) 0–772.8 (n = 119)  −143.3 (202.6) −772.8 to 680.4 (n = 109)  −161.3 (168.5) −907.2 to 165.2 (n = 119)  0.464  Heavy drinking days (per month)  13.1 (8.0) 1–30 (n = 109)  7.9 (8.9) 0–30 (n = 109)  11.9 (8.0) 0–30 (n = 119)  6.8 (7.8) 0–30 (n = 119)  −5.2 (8.9) −30 to 19 (n = 109)  −5.1 (7.8) −30 to 16 (n = 119)  0.954  ICD-10 criteria dependence  4.2 (1.0) 3–6 (n = 109)  2.1 (1.7) 0–6 (n = 105)  4.3 (1.0) 3–6 (n = 119)  2.2 (1.7) 0–6 (n = 118)  −2.1 (1.8) −6 to 2 (n = 105)  −2.1 (1.8) −6 to 3 (n = 118)  0.900  AUDIT total score  22.4 (5.8) 6–37 (n = 106)  14.1 (7.0) 0–32 (n = 100)  23.2 (4.9) 10–33 (n = 115)  13.5 (6.5) 0–35 (n = 111)  −8.4 (5.9) −30 to 3 (n = 99)  −9.3 (7.6) −30 to 7 (n = 107)  0.325  SADD total score  14.4 (6.3) 3–35 (n = 105)  9.0 (5.7) 0–34 (n = 100)  14.2 (5.3) 4–33 (n = 115)  8.0 (4.3) 0–23 (n = 114)  −5.5 (5.2) −20 to 3 (n = 98)  −6.2 (5.8) −23 to 5 (n = 110)  0.392  SIP total score  14.9 (7.0) 2–32 (n = 108)  8.0 (5.8) 0–29 (n = 103)  16.0 (6.5) 3–33 (n = 118)  7.8 (5.6) 0–22 (n = 115)  −6.8 (6.4) −26 to 7 (n = 103)  −7.9 (7.1) −30 to 7 (n = 114)  0.234  HADS scale anxiety  6.9 (4.1) 0–19 (n = 108)  5.7 (3.9) 0–19 (n = 104)  7.3 (4.4) 0–19 (n = 118)  5.6 (3.5) 0–17 (n = 114)  −1.1 (3.3) −9 to 15 (n = 103)  −1.4 (3.8) −14 to 7 (n = 113)  0.535  HADS scale depression  4.3 (3.4) 0–16 (n = 108)  3.1 (2.6) 0–10 (n = 105)  4.6 (3.2) 0–15 (n = 118)  3.1 (2.9) 0–15 (n = 116)  −1.3 (2.8) −9 to 8 (n = 104)  −1.5 (3.4) −12 to 7 (n = 115)  0.629  EQ 5D 5 L VAS  63.2 (18.8) 10–95 (n = 107)  73.7 (16.9) 20–100 (n = 105)  64.9 (18.4) 10–100 (n = 119)  75.5 (15.0) 30–100 (n = 116)  10.4 (19.5) −70 to 55 (n = 103)  10.5 (16.1) −35 to 65 (n = 116)  0.990  CDT  2.5 (2.1) 0.6–10.8 (n = 108)  2.4 (2.0) 0.6–12.5 (n = 99)  2.3 (2.0) 0.9–13.9 (n = 117)  1.9 (1.7) 0.6–13.6 (n = 107)  −0.2 (1.6) −6.9 to 4.6 (n = 98)  −0.28 (1.00) −6.2 to 2.4 (n = 105)  0.668  GT  0.8 (1.4) 0.1–11.8 (n = 109)  0.6 (0.9) 0.1–5.6 (n = 99)  0.8 (1.3) 0.1–8.6 (n = 118)  0.6 (1.0) 0.1–7.4 (n = 107)  −0.2 (1.4) −11.3 to 2.7 (n = 99)  −0.2 (0.6) −2.8 to 1.1 (n = 106)  0.710  ASAT  0.5 (0.3) 0.2–1.6 (n = 109)  0.5 (0.4) 0.2–3.6 (n = 98)  0.5 (0.3) 0.2–2.1 (n = 118)  0.5 (0.3) 0.2–1.7 (n = 107)  −0.0 (0.4) −1.0 to 3.1 (n = 98)  −0.1 (0.3) −1.7 to 0.8 (n = 106)  0.808  ALAT  0.6 (0.4) 0.1–2.2 (n = 109)  0.5 (0.4) 0.1–2.7 (n = 99)  0.5 (0.4) 0.1–2.8 (n = 118)  0.5 (0.4) 0.1–2.5 (n = 107)  −0.0 (0.4) −1.7 to 2.3 (n = 99)  0.0 (0.3) −1.9 to 1.6 (n = 106)  0.569  CSQ total score    21.6 (6.0) 9–32 (n = 95)    26.1 (4.8) 12–32 (n = 109)      <0.001*  Proportion below the Swedish national guidelines for hazardous consumptiona [n (%)]  6 (4.2) (n = 109)  19 (17.4) (n = 109)  4 (2.8) (n = 119)  23 (19.3) (n = 119)  −13 (−13.2) (n = 109)  −19 (−16.5)(n = 119)  0.843 df(1)  aHDD = 0 and men ≤14 units/week; women ≤9 units/week (1 unit = 12 g alcohol. The results from the secondary outcomes follow the same pattern as the primary outcome, with no statistically significant differences between the groups. The only exception was client satisfaction, with significant differences in CSQ scores, where patients treated in specialist care were more satisfied with the treatment (t = 5.9, P < 0.001). Predictors for treatment outcome Additional analyses were performed for available cases, investigating whether length of treatment and compliance with treatment affected the primary outcome. As these are results from sub-analyses, caution is warranted in the interpretations. In both treatment conditions a higher number of visits predicted lower weekly alcohol consumption at 6 months (n = 228, −12.69, 95% CI: −21.76; −3.63, P = 0.006). However, length of treatment measured in minutes did not predict the primary outcome (n = 228, −0.16, 95% CI: −0.37; 0.06 P = 0.146). When number of visits were controlled for, the results confirm that primary care was non-inferior to specialist care (n = 228, 4.73, 95% CI: −38.71; 48.18, P = 0.830). Non-inferiority was also confirmed for participants who self rated that they had completed the treatment they were randomized to (n = 144, −5.43, 95% CI: −42.28; 31.42, P = 0.771). Analyses were also performed to investigate whether participants’ gender, age, severity of dependence, level of anxiety or depression and readiness to change, affected the primary outcome weekly alcohol consumption (grams of alcohol) in primary care compared to specialist care (Fig. 2). Fig. 2. View largeDownload slide Sub-analyses of predictors for differences in primary outcome in primary care compared to specialist treatment. ITT data set with available cases. The value on the coefficient indicates the estimated mean difference between primary care and specialist care. Fig. 2. View largeDownload slide Sub-analyses of predictors for differences in primary outcome in primary care compared to specialist treatment. ITT data set with available cases. The value on the coefficient indicates the estimated mean difference between primary care and specialist care. Participants with AUDIT scores indicative of a low severity of dependence (<20), showed a significantly larger reduction of weekly alcohol consumption in primary care compared to specialist care. Women and participants with low severity of dependence irrespective of gender, measured with SADD, showed non-inferior outcomes in primary care compared with specialist care, as the high ends of the confidence intervals were below 50 g of alcohol. On the other hand, patients with a high severity of alcohol dependence, measured with AUDIT or SADD, or high levels of depression showed a significantly larger reduction in alcohol consumption when treated in specialist care. There was no difference in treatment outcome in primary care compared to specialist care for younger participants, compared to middle aged or older. Regarding readiness to change, a higher rating of confidence predicted a lower weekly alcohol consumption at 6 months (n = 219, −1.38, 95% CI: −2.40; −0.37, P = 0.007). Importance did not predict alcohol consumption at 6 months (n = 220, −0.75, 95% CI: −2.80; 1.30, P = 0.474). The readiness to change measures did not alter the estimate for the difference between primary care and specialist care on the primary outcome measure (n = 219, 27.63, 95% CI: −12.94; 68.19, P = 0.181) Lost to follow up The follow-up rate was 81% overall. In the primary care condition 24% (n = 34) were lost to follow up and 15% (n = 22) in specialist care. However, this is not statistically significant (χ2 = 1.705, df = 1, P = 0.19). Participants randomized to primary care who were lost to follow up had significantly lower weekly alcohol consumption at baseline compared to those who completed followed up (mean 269 respectively 367, t = 2.49, P = 0.014). They also had a lower consumption compared to participants lost to follow up after specialist care (mean 377, t = 2.25, P = 0.028). In both treatment conditions, participants lost to follow up were younger (age mean 50 respectively 56, t = 4.00, P = 0.000), had higher levels of depression at baseline (HADS depression mean 6.1 respectively 4.8, t = 3.42, P = 0.001), and higher levels of anxiety at baseline (HADS anxiety mean 9.0 respectively 7.1, t = 3.00, P = 0.003), compared to those who completed followed up. Side effects At follow up, seven participants from primary care and six from specialist care, reported side effects from the pharmacological treatment (one could report more than one side effect): nausea (n = 4); diarrhoea (n = 2); anxiety (n = 2); thrombocytopenia (n = 1); perspirations (n = 1); shaky (n = 1); stomach pain (n = 1); sleeping difficulties (n = 1); tiredness (n = 1); and vertigo (n = 1). Six participants did not specify which type of side effects. One patient, who was randomized to primary care, died during the study period (cause unknown). DISCUSSION Overall findings The aim of this study was to evaluate if treatment of alcohol dependence in primary care, delivered by general practitioners, is equally effective as treatment by specialists in reducing alcohol consumption. The results for the primary outcome were statistically inconclusive, where we cannot conclude that primary care treatment is non-inferior to specialist care. On the other hand, treatment in specialist care was not found to be superior to treatment in primary care. Only on satisfaction with treatment, was specialist care superior to primary care. However, the results from sub-analysis are very informative. Primary care treatment was non-inferior to specialist care among participants who reported that they completed the treatment. The general practitioners only received 1 day of training, and no further supervision during the trial. It is possible that more training or regular supervision could have improved the general practitioners' skills, and thereby improved retention to treatment for the participants. This study also illustrates the role of severity of dependence for treatment outcome in different settings. Sub-analyses showed that individuals with alcohol dependence of low severity reduced their consumption to the same extent in primary care as in specialist care. In contrast, individuals with alcohol dependence of high severity showed better outcomes after specialist treatment. Also, patients with higher level of depression showed better treatment outcomes in specialist care. Previous studies of treatment in primary care have shown varied results (Oslin et al., 2013; O’Malley et al., 2003; Drummond et al., 1990). The manual in this study is briefer compared to other manuals for treatment of alcohol dependence in primary care. In O’Malley et al. (2003) the primary care treatment was an initial 45-min visit followed by seven 15–20 min sessions. While in Oslin et al. (2013) the primary care treatment covered on average ten 30 min visits. As number of visits contributed significantly to the outcome, it is possible the 15-method was too brief to facilitate change among individuals with more severe dependence and higher levels of depression. On the other hand, it appeared sufficient for the larger group with moderate levels of dependence. This suggests that the more severe group should be offered additional treatment or be referred to specialized care. This study included individuals with alcohol dependence who traditionally have been reluctant to seek treatment. Participants showed lower weekly alcohol consumption, fewer psychiatric co-morbidities, higher education and higher rates of co-habiting compared to previous studies on treatment of alcohol dependence in primary care (Oslin et al., 2013; Coste et al., 2016). At baseline, the weekly alcohol consumption in this sample corresponded to the level found among primary care patients with alcohol dependence who are not treatment seekers for alcohol problems (Rehm et al., 2015b). This is possibly because the main recruitment method was through advertisements, which is known to attract individuals with lower problem severity compared to clinical samples (Morley et al., 2009). While the majority of participants were still drinking at hazardous levels at 6 months follow-up, they had reduced their consumption substantially. From a public health perspective, it needs to be stressed that heavy drinking is a strong risk factor for mortality, and that any decrease in consumption is associated with a reduced risk for mortality (Laramee et al., 2015). Limitations This study reports outcomes of primary care treatment for identified alcohol dependent patients. A major challenge in primary care however, is the identification of alcohol problems, which is the first step of the 15-method, but not part of this study. As long as practitioners are unable to identify patients with alcohol problems, having access to effective treatment will only reach individuals who are concerned about their drinking and actively seek primary care treatment. A limitation in the study design lies in the choice of the comparison treatment. The specialized addiction unit was recently developed to be an attractive alternative to treatment as usual in specialized addiction care, for the same segment of the dependence population as the sample of this study. This may have improved the retention to treatment and thereby the outcomes in the specialist treatment condition, compared to if the participants had been randomized to regular specialist care. In order to keep the study as naturalistic as possible, the data collected on treatment fidelity were only self-reported. However, this is also a limitation to the study, as there are no objective ratings of fidelity to the manual. As the participants were alcohol dependent, and consumed alcohol at harmful levels, the design did not, due to ethical reasons, include a no-treatment control. Even though natural recovery is common among individuals with alcohol dependence, seeking treatment improves the rate of recovery, independent of severity of dependence (Dawson et al., 2006). In Dawson et al. (2006), 45.7% of treatment seekers recovered, while the corresponding numbers for non-treatment seekers were 32.5%. It is therefore important to reach people with alcohol dependence with treatments that they find attractive and are willing to seek. Lastly, nearly 20% of the participants were lost to follow up. This level of drop out is correspondent to previous studies, for example UKATT (UKATT, 2005), but nevertheless constitutes an impediment to the conclusions. Generalizability The general practitioners who delivered the interventions were ordinary clinicians, who received 1-day training in the method. The generalizability to clinical settings is therefore very good. The participants had lower severity of problems compared to other studies. From an epidemiological perspective, this represents the majority of individuals with alcohol dependence (Andreasson et al., 2013). However, as the participants’ actively sought treatment by responding to an advertisement, and showed high scores on readiness to change, the results may not be generalized to proactively recruited individuals. Interpretation Primary care practitioners can, compared to addiction specialists, successfully treat individuals with a low to moderate severity of alcohol dependence. The 15-method, may be a way to broaden the base of treatment for alcohol dependence, reducing the treatment gap in this field. By offering services in primary care, it is possible to reach a larger share of all individuals with alcohol dependence. This is especially important in jurisdictions with little or no specialized addiction treatment. ACKNOWLEDGEMENTS We want to thank the participating primary care units and the general practitioners who delivered the treatments. We also want to thank Victoria Andersson, research nurse, who was the study coordinator. CONFLICT OF INTEREST STATEMENT None declared. OTHER INFORMATION The study protocol is available from the corresponding author. REFERENCES Alvarez-Bueno C, Rodriguez-Martin B, Garcia-Ortiz L, et al.  . ( 2015) Effectiveness of brief interventions in primary health care settings to decrease alcohol consumption by adult non-dependent drinkers: a systematic review of systematic reviews. Prev Med  76: S33– 8. Google Scholar CrossRef Search ADS PubMed  Andreasson S, Danielsson AK, Hallgren M. ( 2013) Severity of alcohol dependence in the Swedish adult population: association with consumption and social factors. Alcohol  47: 21– 5. Google Scholar CrossRef Search ADS PubMed  Andreasson S, Hansagi H, Osterlund B. ( 2002) Short-term treatment for alcohol-related problems: four-session guided self-change versus one session of advice—a randomized, controlled trial. Alcohol  28: 57– 62. Google Scholar CrossRef Search ADS PubMed  Babor TF, Higgins-Biddle JC, Saunders JB, et al.  . ( 2001) AUDIT—The Alcohol Use Disorders Identification Test. Guidelines for use in Primary Care . Switzerland: World Health Organization. Berger L, Fisher M, Brondino M, et al.  . ( 2013) Efficacy of acamprosate for alcohol dependence in a family medicine setting in the United States: a randomized, double-blind, placebo-controlled study. Alcohol Clin Exp Res  37: 668– 74. Google Scholar CrossRef Search ADS PubMed  Bischof G, Grothues JM, Reinhardt S, et al.  . ( 2008) Evaluation of a telephone-based stepped care intervention for alcohol-related disorders: a randomized controlled trial. Drug Alcohol Depend  93: 244– 51. Google Scholar CrossRef Search ADS PubMed  Chick J, Ritson B, Connaughton J, et al.  . ( 1988) Advice versus extended treatment for alcoholism: a controlled study. Br J Addict  83: 159– 70. Google Scholar CrossRef Search ADS PubMed  Coste F, Rahhali N, Wallace P. ( 2016) STREAM: elucidating how alcohol use disorders are managed in UK General Practice Today. Alcohol Alcohol . 51: 428– 35. Google Scholar CrossRef Search ADS PubMed  Dawson DA, Grant BF, Stinson FS, et al.  . ( 2006) Estimating the effect of help-seeking on achieving recovery from alcohol dependence. Addiction  101: 824– 34. Google Scholar CrossRef Search ADS PubMed  Drummond DC, Thom B, Brown C, et al.  . ( 1990) Specialist versus general practitioner treatment of problem drinkers. Lancet  336: 915– 8. Google Scholar CrossRef Search ADS PubMed  Glass JE, Hamilton AM, Powell BJ, et al.  . ( 2015) Specialty substance use disorder services following brief alcohol intervention: a meta-analysis of randomized controlled trials. Addiction  110: 1404– 15. Google Scholar CrossRef Search ADS PubMed  Herdman M, Gudex C, Lloyd A, et al.  . ( 2011) Development and preliminary testing of the new five-level version of EQ-5D (EQ-5D-5L). Qual Life Res  20: 1727– 36. Google Scholar CrossRef Search ADS PubMed  Karhuvaara S, Simojoki K, Virta A, et al.  . ( 2007) Targeted nalmefene with simple medical management in the treatment of heavy drinkers: a randomized double-blind placebo-controlled multicenter study. Alcohol Clin Exp Res  31: 1179– 87. Google Scholar CrossRef Search ADS PubMed  Kiritze-Topor P, Huas D, Rosenzweig C, et al.  . ( 2004) A pragmatic trial of acamprosate in the treatment of alcohol dependence in primary care. Alcohol Alcohol  39: 520– 7. Google Scholar CrossRef Search ADS PubMed  Kohn R, Saxena S, Levav I, et al.  . ( 2004) The treatment gap in mental health care. Bull World Health Organ  82: 858– 66. Google Scholar PubMed  Laramee P, Leonard S, Buchanan-Hughes A, et al.  . ( 2015) Risk of all-cause mortality in alcohol-dependent individuals: a systematic literature review and meta-analysis. EBioMedicine  2: 1394– 404. Google Scholar CrossRef Search ADS PubMed  Larsen DL, Attkisson CC, Hargreaves WA, et al.  . ( 1979) Assessment of client/patient satisfaction: development of a general scale. Eval Program Plann  2: 197– 207. Google Scholar CrossRef Search ADS PubMed  Lieberman DZ, Cioletti A, Massey SH, et al.  . ( 2014) Treatment preferences among problem drinkers in primary care. Int J Psychiatry Med  47: 231– 40. Google Scholar CrossRef Search ADS PubMed  Magill M, Ray LA. ( 2009) Cognitive-behavioral treatment with adult alcohol and illicit drug users: a meta-analysis of randomized controlled trials. J Stud Alcohol Drugs  70: 516– 27. Google Scholar CrossRef Search ADS PubMed  Miller WR, Sovereign RG, Krege B. ( 1988) Motivational interviewing with problem drinkers: II. The Drinker’s Check-up as a preventive intervention. Behavioural Psychotherapy  16: 251– 68. Google Scholar CrossRef Search ADS   Miller WR, Tonigan JS, Longabaugh R. ( 1995) The Drinker Inventory of Consequences (DrInC): an instrument for assessing adverse consequences of alcohol abuse (Test Manual). NIAA Project MATCH Monograph Series . Bethesda: : Department of Health and Human Service. Morley KC, Teesson M, Sannibale C, et al.  . ( 2009) Sample bias from different recruitment strategies in a randomised controlled trial for alcohol dependence. Drug Alcohol Rev  28: 222– 9. Google Scholar CrossRef Search ADS PubMed  Oslin DW, Lynch KG, Maisto SA, et al.  . ( 2013) A randomized clinical trial of alcohol care management delivered in Department of Veterans Affairs Primary Care Clinics Versus Specialty Addiction Treatment. J Gen Intern Med  29: 162– 8. Google Scholar CrossRef Search ADS PubMed  O’Donnell A, Anderson P, Newbury-Birch D, et al.  . ( 2014) The impact of brief alcohol interventions in primary healthcare: a systematic review of reviews. Alcohol Alcohol  49: 66– 78. Google Scholar CrossRef Search ADS PubMed  O’Malley SS, Rounsaville BJ, Farren C, et al.  . ( 2003) Initial and maintenance naltrexone treatment for alcohol dependence using primary care vs specialty care: a nested sequence of 3 randomized trials. Arch Intern Med  163: 1695– 704. Google Scholar CrossRef Search ADS PubMed  Probst C, Manthey J, Martinez A, et al.  . ( 2015) Alcohol use disorder severity and reported reasons not to seek treatment: a cross-sectional study in European primary care practices. Subst Abuse Treat Prev Policy  10: 32. Google Scholar CrossRef Search ADS PubMed  Projectmatch. ( 1997) Matching alcoholism treatments to client heterogeneity: project MATCH posttreatment drinking outcomes. J Stud Alcohol  58: 7– 29. CrossRef Search ADS PubMed  Raistrick DD, Davidson GR. ( 1983) Development of a questionnaire to measure alcohol dependence. Br J Addict  78: 89– 95. Google Scholar CrossRef Search ADS PubMed  Rehm J, Allamani A, Elekes Z, et al.  . ( 2015a) Alcohol dependence and treatment utilization in Europe—a representative cross-sectional study in primary care. BMC Fam Pract  16: 90. Google Scholar CrossRef Search ADS PubMed  Rehm J, Manthey J, Struzzo P, et al.  . ( 2015b) Who receives treatment for alcohol use disorders in the European Union? A cross-sectional representative study in primary and specialized health care. Eur Psychiatry  30: 885– 93. Google Scholar CrossRef Search ADS PubMed  Rehm J, Shield KD, Gmel G, et al.  . ( 2013) Modeling the impact of alcohol dependence on mortality burden and the effect of available treatment interventions in the European Union. Eur Neuropsychopharmacol  23: 89– 97. Google Scholar CrossRef Search ADS PubMed  Riddargatan 1 ( 2012). Manual för 15-metoden (Swedish). http://www.ridddargatan1.se/PV/: Riddargatan 1, Stockholm Centre for Dependency Disorders/Centre for Psychiatry Research, Stockholm. Saitz R. ( 2010) Alcohol screening and brief intervention in primary care: absence of evidence for efficacy in people with dependence or very heavy drinking. Drug Alcohol Rev  29: 631– 40. Google Scholar CrossRef Search ADS PubMed  Saitz R, Cheng DM, Winter M, et al.  . ( 2013) Chronic care management for dependence on alcohol and other drugs: the AHEAD randomized trial. J Am Med Assoc  310: 1156– 67. Google Scholar CrossRef Search ADS   Sellman JD, Sullivan PF, Dore GM, et al.  . ( 2001) A randomized controlled trial of motivational enhancement therapy (MET) for mild to moderate alcohol dependence. J Stud Alcohol  62: 389– 96. Google Scholar CrossRef Search ADS PubMed  Sobell MB, Sobell LC. ( 1993) Problem Drinkers: Guided Self-Change Treatment . New York: Guilford. Sobell LC, Sobell MB 1996. Time Line Follow Back. User´s Guide, Toronto, Addiction Research Foundation. Storbjork J, Room R. ( 2008) The two world of alcohol problems: Who is in treatment and who is not? Addiction Research & Theory  16: 67– 84. Google Scholar CrossRef Search ADS   The World Health Organization. ( 1992) The ICD-10 Classification of Mental and Behavioural Disorders . Geneva: World Health Organization. Trim RS, Schuckit MA, Smith TL. ( 2013) Predictors of initial and sustained remission from alcohol use disorders: findings from the 30-year follow-up of the San Diego Prospective Study. Alcohol Clin Exp Res  37: 1424– 31. Google Scholar CrossRef Search ADS PubMed  Ukatt. ( 2005) Effectiveness of treatment for alcohol problems: findings of the randomised UK alcohol treatment trial (UKATT). Br Med J  331: 541. CrossRef Search ADS   Upshur C, Weinreb L, Bharel M, et al.  . ( 2015) A randomized control trial of a chronic care intervention for homeless women with alcohol use problems. J Subst Abuse Treat  51: 19– 29. Google Scholar CrossRef Search ADS PubMed  Wallhed Finn S, Bakshi A-S, Andréasson S ( 2014). Alcohol consumption, dependence, and treatment barriers: perceptions among nontreatment seekers with alcohol dependence. Substance Use & Misuse 49: 762– 9. Walters GD. ( 2000) Behavioral self-control training for problem drinkers: a meta-analysis of randomized control studies. Behav Ther  31: 135– 49. Google Scholar CrossRef Search ADS   Watkins KE, Ober AJ, Lamp K, et al.  . ( 2017) Collaborative care for opioid and alcohol use disorders in primary care: The SUMMIT Randomized Clinical Trial. JAMA Intern Med  177: 1480– 88. Google Scholar CrossRef Search ADS PubMed  Willenbring ML, Olson DH. ( 1999) A randomized trial of integrated outpatient treatment for medically ill alcoholic men. Arch Intern Med  159: 1946– 52. Google Scholar CrossRef Search ADS PubMed  Witkiewitz K, Falk DE, Kranzler HR, et al.  . ( 2014) Methods to analyze treatment effects in the presence of missing data for a continuous heavy drinking outcome measure when participants drop out from treatment in alcohol clinical trials. Alcohol Clin Exp Res  38: 2826– 34. Google Scholar CrossRef Search ADS PubMed  Zigmond AS, Snaith RP. ( 1983) The hospital anxiety and depression scale. Acta Psychiatr Scand  67: 361– 70. Google Scholar CrossRef Search ADS PubMed  Author notes Clinical Trial Registration: Trial identifier ISRCTN84490505 at Controlled Trials.com, assigned 2013-07-09. © The Author(s) 2018. Medical Council on Alcohol and Oxford University Press. All rights reserved. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Alcohol and Alcoholism Oxford University Press

Treatment for Alcohol Dependence in Primary Care Compared to Outpatient Specialist Treatment—A Randomized Controlled Trial

Loading next page...
 
/lp/ou_press/treatment-for-alcohol-dependence-in-primary-care-compared-to-8RKlSCBAEU
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Medical Council on Alcohol and Oxford University Press. All rights reserved.
ISSN
0735-0414
eISSN
1464-3502
D.O.I.
10.1093/alcalc/agx126
Publisher site
See Article on Publisher Site

Abstract

Abstract Aim To investigate if treatment for alcohol dependence in primary care is as effective as specialist addiction care. Method Randomized controlled non-inferiority trial, between groups parallel design, not blinded. The non-inferiority limit was set to 50 grams of alcohol per week. About 288 adults fulfilling ICD-10 criteria for alcohol dependence were randomized to treatment in primary care (men n = 82, women n = 62) or specialist care (men n = 77, women n = 67). General practitioners at 12 primary care centers received 1-day training in a treatment manual for alcohol dependence. Primary outcome was change in weekly alcohol consumption at 6-months follow-up compared with baseline, as measured with timeline follow back. Secondary outcomes were heavy drinking days, severity of dependence, consequences of drinking, psychological health, quality of life, satisfaction with treatment and biomarkers. Results Intention-to-treat analysis (n = 228) was statistically inconclusive, and could not confirm non-inferiority for the primary outcome, since the high end of the confidence interval exceeded 50 grams (estimated mean weekly alcohol consumption was 30 grams higher in primary care compared with specialist care; 95% confidence interval −10.20; 69.72). However, treatment in specialist care was not significantly superior to primary care (P = 0.146). Subanalysis suggests that specialist care was superior to primary care only for patients with high severity of dependence. Conclusions Treatment for alcohol dependence in primary care is a promising approach, especially for individuals with low to moderate dependence. This may be a way to broaden the base of treatment for alcohol dependence, reducing the current treatment gap. INTRODUCTION Though treatment seeking has been found to increase the rates of recovery from alcohol dependence (Trim et al., 2013) fewer than one in five seek treatment (Rehm et al., 2015a). Alcohol dependence has one of the largest treatment gaps between the number of individuals affected and the number in treatment (Kohn et al., 2004). Treatment seekers typically have a more severe form of dependence, often in combination with co-morbid disorders and a more unstable social situation, e.g. homelessness or unemployment (Kohn et al., 2004; Storbjork and Room, 2008). In addition, individuals with moderate alcohol dependence compared to more severe forms of dependence tend to have less psychiatric co-morbidities, a more stable social situation and more rarely seek treatment. Several treatments have shown efficacy in reducing alcohol consumption for individuals with dependence. In the EU, estimates show that 12,000 lives could annually be saved by treating two out of five alcohol dependent individuals, rather than today's one in five (Rehm et al., 2013). Thus, there is a need for the treatment services to reach and treat a larger proportion of individuals with excessive alcohol consumption. An important barrier to seek treatment is the stigma surrounding alcohol dependence and treatment seeking (Wallhed Finn et al., 2014; Probst et al., 2015). One possible way of reducing the stigma is to offer treatment for alcohol dependence in primary care. Studies indicate that individuals with alcohol dependence are positive to seeking treatment in primary care (Lieberman et al., 2014; Wallhed Finn et al., 2014). In addition, a large proportion of individuals with alcohol dependence are already in primary care treatment for other conditions (Rehm et al., 2015a). Research on interventions for alcohol problems in primary care have mainly focused on Screening, Brief Interventions and Referral to Treatment, also known as SBIRT (O’Donnell et al., 2014). However, there is a lack of evidence that screening and brief interventions lead to a higher rate of treatment utilization in specialist care among problem drinkers, independent of problem severity (Glass et al., 2015). Most treatment studies in primary care have included individuals with hazardous or harmful consumption but alcohol dependent patients have been studied to a lesser extent. A British study found no difference in reduction of alcohol consumption for patients with alcohol dependence treated by general practitioners compared to specialist treatment (Drummond et al., 1990). However, the general practitioners received continuous support from the specialists during the course of treatments. There was also a significant cross over between the treatment arms. A stepped care intervention delivered via telephone has been shown to reduce alcohol consumption among at-risk drinkers in primary care (Bischof et al., 2008). However, no effects were found among individuals with alcohol dependence. Longitudinal care with Chronic Care Management in primary care, targeting individuals with severe dependence, has shown mixed results compared to treatment as usual in primary care (Willenbring and Olson, 1999; Saitz et al., 2013; Upshur et al., 2015). A recent study has shown that care based on similar principles increased treatment utilization and abstinence among individuals with opioid and alcohol dependence (Watkins et al., 2017). Two studies have investigated treatment with naltrexone combined with behavioural interventions in primary care compared to specialist care (O’Malley et al., 2003; Oslin et al., 2013). In both studies, the behavioural interventions were delivered by generalists without specialist training in treatment for alcohol dependence, and the caregivers were regularly supervised. O’Malley et al. (2003) found an advantage for specialist care on percentage of days abstinent, but no differences on secondary drinking outcomes. Oslin et al. (2013) found treatment in primary care more effective in reducing heavy drinking days (HDD) compared to specialist care. Moreover, retention to treatment was significantly higher in primary care. Three studies have evaluated the effects of pharmacological treatment in primary care delivered by generalists without specialist training. Kiritze-Topor et al. (2004) showed that adding acamprosate to treatment as usual, resulted in a longer duration of abstinence. However, a study from North Carolina and Wisconsin showed no significant effects of acamprosate compared to placebo, when delivered concomitant to five sessions of brief intervention (Berger et al., 2013). As-needed nalmefene, together with a minimal psychosocial intervention, was found to reduce the number of HDD compared to placebo (Karhuvaara et al., 2007). Thus, there is preliminary evidence for successful treatment of alcohol dependence by generalists in primary care. However, as time constraints are an important issue in primary care, one important challenge is to develop brief and effective treatment models that are feasible to implement. Also, in regular clinical practice, access to supervision from specialists is scarce, emphasizing the need of evaluating interventions delivered by generalists without supervision. As stand-alone interventions, brief interventions have been shown to reduce alcohol consumption among hazardous drinkers, but not individuals with alcohol dependence (Saitz, 2010). We have developed a manual for treatment of hazardous drinking and alcohol dependence in primary care, the 15-method. The manual starts with a brief intervention and, if needed, the patient proceeds to interventions that are more extensive with the possibility of receiving pharmacological and brief psychological treatment (Riddargatan 1, 2012). The model consists of three steps: (1) identification of problem drinking and brief advice (Alvarez-Bueno et al., 2015); (2) assessment, with a 30-min feedback session (Chick et al., 1988; Miller et al., 1988); and (3) four sessions based on CBT and motivational interviewing (Sobell and Sobell, 1993; Andreasson et al., 2002). Each session contains a theme to facilitate behaviour change: goal setting, self-monitoring of alcohol consumption, identifying risk situations and problem solving. These sessions can be combined with pharmacological treatment; acamprosate, disulfiram, nalmefene or naltrexone. The name, ‘the 15-method’ refers to the length of sessions, 15 min, and that the final two steps of the intervention target patients who score >15 points on the AUDIT (Babor et al., 2001). Shared decision-making guides the type of treatment, treatment goal and intensity of treatment provided. We have found no study of brief treatment, integrated with the choice of the full range of available pharmacological treatment delivered in primary care, adapted to the context and time constraints of primary care. Furthermore, to our knowledge there are no studies of treatment for alcohol dependence delivered by general practitioners without access to continuous support and supervision from specialists. Objectives The overall objective of the present study was to investigate if treatment for alcohol dependence in primary care was as effective as standard treatment delivered in specialist care. We will present results from a qualitative study, where practitioners and managers at the primary health care centres involved in this study were interviewed, in a separate publication. Hypothesis Treatment for alcohol dependence in primary care with the 15-method was non-inferior to specialist care. METHODS The study was approved by the regional ethics board in Stockholm, 2012-11-07, ref: 2012/1760-31/1. Trial design Randomized controlled non-inferiority trial, between groups parallel design 1:1 allocation ratio. Participants Participants were recruited via advertisement in newspapers, leaflets at the primary care study sites or direct question at a primary care consultation. After an initial telephone screening with a research nurse, eligible individuals, who preliminarily fulfilled the ICD-10 criteria for alcohol dependence, were scheduled for an assessment at the research unit. At the face-to-face assessment, a more detailed assessment of the ICD-10 criteria was conducted. Standardized protocols for both the telephone and the face-to-face assessments were used. The research nurse was trained by the third author in conducting the assessment protocol. The inclusion criteria were as follows: alcohol dependence according to ICD-10; ≥18 years old and living within Stockholm county. The exclusion criteria were as follows: need of continuous support from the social services (homeless, financial support); previous severe withdrawal symptoms (seizures, hallucinations, delirium); misuse or dependence of narcotics or prescription drugs (ICD-10); severe somatic or psychiatric condition; or not able to understand Swedish. Interventions Primary care The manual was adapted from the 15-method (Riddargatan 1, 2012), which is a model for treatment of hazardous drinking and alcohol dependence in primary care. For this study, where participants were mainly recruited by advertisement, and had already completed intake assessment as a part of study procedures, the first step of the manual was redundant. Shared decision-making between the patient and physician guided whether participants received step 3 and which type of treatment (psychological, pharmacological or a combination), and also the treatment goal; abstinence or controlled drinking. The treatment according to the manual consisted of a maximum of five sessions or 90 min over 24 weeks. Specialist treatment Participants randomized to specialist treatment, first received feedback of the baseline assessment, delivered by a physician. Participants were then offered pharmacological and/or psychological treatment, and shared decision-making also guided the choice of treatment. The pharmacological treatments were identical to the ones in primary care regarding substances and dosage, however, with 30 min per visit. The manual based psychological treatments were either motivational enhancement treatment (four sessions) (Sellman et al., 2001), guided self-change (four sessions) (Sobell and Sobell, 1993), relapse prevention (six to eight sessions) (Magill and Ray, 2009), behavioural self-control training (four to five sessions) (Walters, 2000) or 12-step treatment (12 sessions) (Project MATCH, 1997). Each session in the psychological treatments lasted 45 min. All staff in both condition indicated the grade of fidelity to the treatment manuals by self-report at the end of each session. Settings The interventions in primary care were delivered by the regular general practitioners, who received an eight hour training in the 15-method. The training consisted of lectures on the manual, combined with skills training including role-play of cases. In total, 29 general practitioners at 12 different primary care centres delivered the treatment. Specialist care was delivered by the clinical staff at Riddargatan 1, an alcohol treatment clinic within the Stockholm Centre for Dependence Disorders. The specialist care was delivered by physicians, nurses, social workers, psychotherapists, psychologists and a 12-step therapist, in total 16 different staff members all with at least several years’ experience in treating alcohol dependence. Outcomes The primary outcome was change in weekly alcohol consumption measured in grams of alcohol before inclusion compared to 6 months after start of treatment. The secondary outcomes were: HDD (women > 3/men > 4 standard units of 12 g of alcohol) per month; severity of dependence; consequences of drinking; symptoms of anxiety and depression; health related quality of life; satisfaction with treatment and biomarkers. An independent research nurse performed baseline and 6-month follow-up measures with the participants. Measurements Questionnaires Time Line Follow Back 30 days (Sobell and Sobell, 1996) was used to assess weekly alcohol consumption and HDD. Severity of alcohol dependence was assessed with Alcohol Use Disorders Identification Test (AUDIT) (Babor et al., 2001); the ICD-10 criteria for alcohol dependence (The World Health Organization, 1992) and Short Alcohol Dependence Data Questionnaire (SADD) (Raistrick and Davidson, 1983). As AUDIT and the ICD-10 criteria assess the last 12 months, these questionnaires were re-worded for the 6-month follow up to only cover the last 6 months. The Short Index of Problems (SIP) assessed consequences of drinking (Miller et al., 1995). Symptoms of anxiety and depression were assessed with Hospital Anxiety and Depression Scale (HADS) (Zigmond and Snaith, 1983). The EQ 5D-5L questionnaire measured health related quality of life (Herdman et al., 2011). Satisfaction with treatment was assessed with Client Satisfaction Questionnaire (CSQ) (Larsen et al., 1979). Readiness to change was assessed with a visual analogue scale (range: 0–100), in which importance and confidence to reach your goal [regarding alcohol] were rated. Participants indicated whether they completed the treatment they were randomized to. Biomarkers Blood was analysed for levels of carbohydrate-deficient transferrin (CDT); gamma-glutamyl transferase (GGT); aspartate amino transferase (AST) and alanine amino transferase (ALT). Non-inferiority limits On the pre-specified primary outcome (difference in weekly alcohol consumption at baseline compared to 6 months following start of treatment), the non-inferiority limit was set to 50 g of alcohol per week. This limit was derived from previous research in our group (Andreasson et al., 2002). In this study, socially well-adjusted heavy drinkers, with moderate levels of dependence severity, received either four sessions of guided self-change or one session of advice. A difference in alcohol consumption of 50 g, or approximately four standard drinks, between groups, was deemed clinically non-significant in this heavy drinking population. Sample size Sample size was calculated for the primary outcome. With a non-inferiority limit set at 50 g of alcohol per week, assuming a standard deviation of 140 g, based on Andreasson et al., 2002, and a difference between groups of 0, a total of 250 patients were necessary to accomplish 80% power to confirm non-inferiority at a 95% confidence level. Assuming a dropout rate of around 20%, the aim was to include 300 participants. Randomization Participants who gave informed consent and completed intake measures were randomized to treatment either at primary care or specialist care. The allocation sequence was generated by the first author, using a computer program (Excel 2010). The sequence was in blocks of eight, and stored with restricted access only by administrative staff not involved in the study. Participants randomized to primary care chose the primary care centre that was geographically most convenient for them. Blinding Neither participants, nor researchers, nor treatment staffs were blinded to the allocations. The study design as a non-inferiority study was known among participants, care providers and outcome assessors. Statistical methods The primary analysis was based on the intention-to-treat (ITT) data set, using data from all available cases as randomized, i.e. all participants who completed follow-up measures at 6 months regardless of protocol adherence. The analysis of the primary outcome was performed using an analysis of covariance (ANCOVA) with the consumption at 6 months as the dependent variable, group as a factor variable, adjusting for baseline consumption. The same statistical model was used for analyses of secondary outcomes. Factors related to the primary outcome was analysed with a regression model. Additional analyses used t-tests and chi-square tests. All analyses used two-sided tests and a significance level of P < 0.05. In-group effect sizes were calculated with Cohen's d. Two sensitivity analyses were performed on the primary outcome. In the first, the ITT data set was used with multiple imputation by predictive mean matching (Witkiewitz et al., 2014). The second analysis was based on a per protocol (PP) principle, where participants with at least one treatment visit, no concurrent treatment for alcohol dependence during the follow up period and who completed the follow-up measures were included. All statistical analyses were conducted with R, Version 3.1.0. RESULTS Participant flow The CONSORT flow chart is shown in Fig. 1. Fig. 1. View largeDownload slide Flow diagram of study enrolment, allocation, 6-month follow-up and analysis. Fig. 1. View largeDownload slide Flow diagram of study enrolment, allocation, 6-month follow-up and analysis. Participants were enroled between October 2013 and March 2015. Due to time constraint, data collection ceased when 288 of the targeted 300 participants had been included. One participant was excluded after randomization to primary care but before first visit due to high liver enzymes. The 6-month follow-up period lasted between May 2014 and January 2016. Table 1 shows demographic and clinical characteristics of each group. Table 1. Baseline characteristics for participants in primary care and specialist care (n = 288)   Primary care (n = 144)  Specialist care (n = 144)  P-value  Gender [n (%)]      0.636   Male  82 (57)  77 (53)     Female  62 (43)  67 (47)    Age [mean (SD) range]  56 (11) 23–77  54 (12) 25–79  0.094  Education [n (%)]      0.031   Not completed compulsory education  2 (1)  0 (0)     9 Years of education  13 (9)  4 (3)     12 Years of education  50 (35)  59 (41)     Higher education  78 (54)  76 (53)     Other  1 (1)  5 (3)    Source of income [n (%)]      1.000   Employment  105 (74)  104 (73)     Pension  31 (22)  31 (22)     Other  6 (4)  7 (5)    Civil status [n (%)]      0.223   Married/co-habiting  83 (58)  92 (64)     Live alone  58 (41)  48 (34)     Widowed  2 (1)  3 (2)    Use of nicotine (current) [n (%)]      0.737   Cigarettes  24 (17)  28 (19)     Snuff  24 (17)  20 (14)     Cigarettes and snuff  7 (5)  10 (7)    Use of illicit drugs (lifetime) [n (%)]  60 (42)  60 (42)  1.000  Problematic alcohol use (years) [mean (SD) range]  11.4 (10.5) 1–55  11.5 (10.7) 0.1–50  0.911  Treatment naive [% (n)]  104 (75)  113 (80)  0.337  Weekly alcohol consumption (g) [mean (SD) range]  343.62 (206.58) 0–1173.2  349.16 (216.45) 0–1680  0.824  Heavy drinking days per month [mean (SD) range]  12.23 (7.81) 0–30  11.92 (7.98) 0–30  0.737  ICD-10 criteria dependence [mean (SD) range]  4.22 (1.01) 2–6  4.39 (1.05) 3–6  0.169  AUDIT total score [mean (SD) range]  22.57 (5.75) 6–37 (n = 141)  23.49 (4.97) 10–38 (n = 138)  0.152  SADD total score [mean (SD) range]  14.83 (6.33) 3–40 (n = 140)  14.39 (5.23) 4–33 (n = 137)  0.535  SIP total score [mean (SD) range]  15.64 (7.48) 2–38 (n = 142)  16.21 (6.64) 3–34 (n = 143)  0.497  HADS scale anxiety [mean (SD) range]  7.36 (4.26) 0–19 (n = 143)  7.58 (4.45) 0–19 (n = 143)  0.674  HADS scale depression [mean (SD) range]  4.71 (3.52) 0–16 (n = 143)  4.84 (3.34) 0–15 (n = 143)  0.743  EQ 5D 5 L VAS [mean (SD) range]  63.11 (17.98) 10–95 (n = 142)  64.49 (18.12) 10–100 (n = 144)  0.518  CDT [mean (SD) range]  2.45 (2.13) 0.13–13.3 (n = 142)  2.28 (1.84) 0.7–13.9 (n = 142)  0.464  GT [mean (SD) range]  0.79 (1.33) 0.1–11.8 (n = 143)  0.86 (1.43) 0.1–9.4 (n = 143)  0.639  ASAT [mean (SD) range]  0.54 (0.35) 0.22–2.8 (n = 143)  0.52 (0.35) 0.22–2.72 (n = 143)  0.616  ALAT [mean (SD) range]  0.57 (0.39) 0.1–2.22 (n = 143)  0.54 (0.42) 0.11–2.73 (n = 143)  0.583  Alcohol consumption below the Swedish national guidelines for hazardous consumption (HDD = 0 and men ≤14 units/week; women ≤9 units/week (1 unit = 12 g alcohol) [n (%)])  6 (4.17)  4 (2.78)  0.520  Readiness to change (importance)  93.60 (9.67) 60–100 (n = 138)  91.88 (11.56) 50–100 (n = 141)  0.179  Readiness to change (confidence)  77.91 (19.99) 0–100 (n = 137)  76.68 (21.03) 5–100 (n = 141)  0.620    Primary care (n = 144)  Specialist care (n = 144)  P-value  Gender [n (%)]      0.636   Male  82 (57)  77 (53)     Female  62 (43)  67 (47)    Age [mean (SD) range]  56 (11) 23–77  54 (12) 25–79  0.094  Education [n (%)]      0.031   Not completed compulsory education  2 (1)  0 (0)     9 Years of education  13 (9)  4 (3)     12 Years of education  50 (35)  59 (41)     Higher education  78 (54)  76 (53)     Other  1 (1)  5 (3)    Source of income [n (%)]      1.000   Employment  105 (74)  104 (73)     Pension  31 (22)  31 (22)     Other  6 (4)  7 (5)    Civil status [n (%)]      0.223   Married/co-habiting  83 (58)  92 (64)     Live alone  58 (41)  48 (34)     Widowed  2 (1)  3 (2)    Use of nicotine (current) [n (%)]      0.737   Cigarettes  24 (17)  28 (19)     Snuff  24 (17)  20 (14)     Cigarettes and snuff  7 (5)  10 (7)    Use of illicit drugs (lifetime) [n (%)]  60 (42)  60 (42)  1.000  Problematic alcohol use (years) [mean (SD) range]  11.4 (10.5) 1–55  11.5 (10.7) 0.1–50  0.911  Treatment naive [% (n)]  104 (75)  113 (80)  0.337  Weekly alcohol consumption (g) [mean (SD) range]  343.62 (206.58) 0–1173.2  349.16 (216.45) 0–1680  0.824  Heavy drinking days per month [mean (SD) range]  12.23 (7.81) 0–30  11.92 (7.98) 0–30  0.737  ICD-10 criteria dependence [mean (SD) range]  4.22 (1.01) 2–6  4.39 (1.05) 3–6  0.169  AUDIT total score [mean (SD) range]  22.57 (5.75) 6–37 (n = 141)  23.49 (4.97) 10–38 (n = 138)  0.152  SADD total score [mean (SD) range]  14.83 (6.33) 3–40 (n = 140)  14.39 (5.23) 4–33 (n = 137)  0.535  SIP total score [mean (SD) range]  15.64 (7.48) 2–38 (n = 142)  16.21 (6.64) 3–34 (n = 143)  0.497  HADS scale anxiety [mean (SD) range]  7.36 (4.26) 0–19 (n = 143)  7.58 (4.45) 0–19 (n = 143)  0.674  HADS scale depression [mean (SD) range]  4.71 (3.52) 0–16 (n = 143)  4.84 (3.34) 0–15 (n = 143)  0.743  EQ 5D 5 L VAS [mean (SD) range]  63.11 (17.98) 10–95 (n = 142)  64.49 (18.12) 10–100 (n = 144)  0.518  CDT [mean (SD) range]  2.45 (2.13) 0.13–13.3 (n = 142)  2.28 (1.84) 0.7–13.9 (n = 142)  0.464  GT [mean (SD) range]  0.79 (1.33) 0.1–11.8 (n = 143)  0.86 (1.43) 0.1–9.4 (n = 143)  0.639  ASAT [mean (SD) range]  0.54 (0.35) 0.22–2.8 (n = 143)  0.52 (0.35) 0.22–2.72 (n = 143)  0.616  ALAT [mean (SD) range]  0.57 (0.39) 0.1–2.22 (n = 143)  0.54 (0.42) 0.11–2.73 (n = 143)  0.583  Alcohol consumption below the Swedish national guidelines for hazardous consumption (HDD = 0 and men ≤14 units/week; women ≤9 units/week (1 unit = 12 g alcohol) [n (%)])  6 (4.17)  4 (2.78)  0.520  Readiness to change (importance)  93.60 (9.67) 60–100 (n = 138)  91.88 (11.56) 50–100 (n = 141)  0.179  Readiness to change (confidence)  77.91 (19.99) 0–100 (n = 137)  76.68 (21.03) 5–100 (n = 141)  0.620  Table 1. Baseline characteristics for participants in primary care and specialist care (n = 288)   Primary care (n = 144)  Specialist care (n = 144)  P-value  Gender [n (%)]      0.636   Male  82 (57)  77 (53)     Female  62 (43)  67 (47)    Age [mean (SD) range]  56 (11) 23–77  54 (12) 25–79  0.094  Education [n (%)]      0.031   Not completed compulsory education  2 (1)  0 (0)     9 Years of education  13 (9)  4 (3)     12 Years of education  50 (35)  59 (41)     Higher education  78 (54)  76 (53)     Other  1 (1)  5 (3)    Source of income [n (%)]      1.000   Employment  105 (74)  104 (73)     Pension  31 (22)  31 (22)     Other  6 (4)  7 (5)    Civil status [n (%)]      0.223   Married/co-habiting  83 (58)  92 (64)     Live alone  58 (41)  48 (34)     Widowed  2 (1)  3 (2)    Use of nicotine (current) [n (%)]      0.737   Cigarettes  24 (17)  28 (19)     Snuff  24 (17)  20 (14)     Cigarettes and snuff  7 (5)  10 (7)    Use of illicit drugs (lifetime) [n (%)]  60 (42)  60 (42)  1.000  Problematic alcohol use (years) [mean (SD) range]  11.4 (10.5) 1–55  11.5 (10.7) 0.1–50  0.911  Treatment naive [% (n)]  104 (75)  113 (80)  0.337  Weekly alcohol consumption (g) [mean (SD) range]  343.62 (206.58) 0–1173.2  349.16 (216.45) 0–1680  0.824  Heavy drinking days per month [mean (SD) range]  12.23 (7.81) 0–30  11.92 (7.98) 0–30  0.737  ICD-10 criteria dependence [mean (SD) range]  4.22 (1.01) 2–6  4.39 (1.05) 3–6  0.169  AUDIT total score [mean (SD) range]  22.57 (5.75) 6–37 (n = 141)  23.49 (4.97) 10–38 (n = 138)  0.152  SADD total score [mean (SD) range]  14.83 (6.33) 3–40 (n = 140)  14.39 (5.23) 4–33 (n = 137)  0.535  SIP total score [mean (SD) range]  15.64 (7.48) 2–38 (n = 142)  16.21 (6.64) 3–34 (n = 143)  0.497  HADS scale anxiety [mean (SD) range]  7.36 (4.26) 0–19 (n = 143)  7.58 (4.45) 0–19 (n = 143)  0.674  HADS scale depression [mean (SD) range]  4.71 (3.52) 0–16 (n = 143)  4.84 (3.34) 0–15 (n = 143)  0.743  EQ 5D 5 L VAS [mean (SD) range]  63.11 (17.98) 10–95 (n = 142)  64.49 (18.12) 10–100 (n = 144)  0.518  CDT [mean (SD) range]  2.45 (2.13) 0.13–13.3 (n = 142)  2.28 (1.84) 0.7–13.9 (n = 142)  0.464  GT [mean (SD) range]  0.79 (1.33) 0.1–11.8 (n = 143)  0.86 (1.43) 0.1–9.4 (n = 143)  0.639  ASAT [mean (SD) range]  0.54 (0.35) 0.22–2.8 (n = 143)  0.52 (0.35) 0.22–2.72 (n = 143)  0.616  ALAT [mean (SD) range]  0.57 (0.39) 0.1–2.22 (n = 143)  0.54 (0.42) 0.11–2.73 (n = 143)  0.583  Alcohol consumption below the Swedish national guidelines for hazardous consumption (HDD = 0 and men ≤14 units/week; women ≤9 units/week (1 unit = 12 g alcohol) [n (%)])  6 (4.17)  4 (2.78)  0.520  Readiness to change (importance)  93.60 (9.67) 60–100 (n = 138)  91.88 (11.56) 50–100 (n = 141)  0.179  Readiness to change (confidence)  77.91 (19.99) 0–100 (n = 137)  76.68 (21.03) 5–100 (n = 141)  0.620    Primary care (n = 144)  Specialist care (n = 144)  P-value  Gender [n (%)]      0.636   Male  82 (57)  77 (53)     Female  62 (43)  67 (47)    Age [mean (SD) range]  56 (11) 23–77  54 (12) 25–79  0.094  Education [n (%)]      0.031   Not completed compulsory education  2 (1)  0 (0)     9 Years of education  13 (9)  4 (3)     12 Years of education  50 (35)  59 (41)     Higher education  78 (54)  76 (53)     Other  1 (1)  5 (3)    Source of income [n (%)]      1.000   Employment  105 (74)  104 (73)     Pension  31 (22)  31 (22)     Other  6 (4)  7 (5)    Civil status [n (%)]      0.223   Married/co-habiting  83 (58)  92 (64)     Live alone  58 (41)  48 (34)     Widowed  2 (1)  3 (2)    Use of nicotine (current) [n (%)]      0.737   Cigarettes  24 (17)  28 (19)     Snuff  24 (17)  20 (14)     Cigarettes and snuff  7 (5)  10 (7)    Use of illicit drugs (lifetime) [n (%)]  60 (42)  60 (42)  1.000  Problematic alcohol use (years) [mean (SD) range]  11.4 (10.5) 1–55  11.5 (10.7) 0.1–50  0.911  Treatment naive [% (n)]  104 (75)  113 (80)  0.337  Weekly alcohol consumption (g) [mean (SD) range]  343.62 (206.58) 0–1173.2  349.16 (216.45) 0–1680  0.824  Heavy drinking days per month [mean (SD) range]  12.23 (7.81) 0–30  11.92 (7.98) 0–30  0.737  ICD-10 criteria dependence [mean (SD) range]  4.22 (1.01) 2–6  4.39 (1.05) 3–6  0.169  AUDIT total score [mean (SD) range]  22.57 (5.75) 6–37 (n = 141)  23.49 (4.97) 10–38 (n = 138)  0.152  SADD total score [mean (SD) range]  14.83 (6.33) 3–40 (n = 140)  14.39 (5.23) 4–33 (n = 137)  0.535  SIP total score [mean (SD) range]  15.64 (7.48) 2–38 (n = 142)  16.21 (6.64) 3–34 (n = 143)  0.497  HADS scale anxiety [mean (SD) range]  7.36 (4.26) 0–19 (n = 143)  7.58 (4.45) 0–19 (n = 143)  0.674  HADS scale depression [mean (SD) range]  4.71 (3.52) 0–16 (n = 143)  4.84 (3.34) 0–15 (n = 143)  0.743  EQ 5D 5 L VAS [mean (SD) range]  63.11 (17.98) 10–95 (n = 142)  64.49 (18.12) 10–100 (n = 144)  0.518  CDT [mean (SD) range]  2.45 (2.13) 0.13–13.3 (n = 142)  2.28 (1.84) 0.7–13.9 (n = 142)  0.464  GT [mean (SD) range]  0.79 (1.33) 0.1–11.8 (n = 143)  0.86 (1.43) 0.1–9.4 (n = 143)  0.639  ASAT [mean (SD) range]  0.54 (0.35) 0.22–2.8 (n = 143)  0.52 (0.35) 0.22–2.72 (n = 143)  0.616  ALAT [mean (SD) range]  0.57 (0.39) 0.1–2.22 (n = 143)  0.54 (0.42) 0.11–2.73 (n = 143)  0.583  Alcohol consumption below the Swedish national guidelines for hazardous consumption (HDD = 0 and men ≤14 units/week; women ≤9 units/week (1 unit = 12 g alcohol) [n (%)])  6 (4.17)  4 (2.78)  0.520  Readiness to change (importance)  93.60 (9.67) 60–100 (n = 138)  91.88 (11.56) 50–100 (n = 141)  0.179  Readiness to change (confidence)  77.91 (19.99) 0–100 (n = 137)  76.68 (21.03) 5–100 (n = 141)  0.620  Length and type of treatment Participants in primary care (n = 144) had on average fewer number of visits compared to specialist care (n = 144), 2.9 (SD = 1.5; range: 0–6) vs 4.7 (SD = 2.8; range: 0–14), (t = 6.913, P < 0.001). On average, participants in primary care (n = 141; missing data n = 3) also received shorter treatment in minutes; 74 (SD = 46; range: 0–195), compared to 187 (SD = 129; range: 0–685) in specialist care (n = 144), (t = 9.88, P = 0.000). In both treatment arms, a combination of pharmacological and psychological treatment was most common. In total, 76 participants in primary care and 62 participants in specialist care received this combination. The number of participants receiving only pharmacological treatment in primary care and in specialist care were 17 and 25, respectively. In general, there were no differences in which type of treatment patients received in the two treatment arms with two exceptions. Firstly, a larger number of patients in specialist care compared to primary care received only psychological treatment, (χ2 = 4.5, df = 1, P= 0.034). Secondly, a larger number of patients in specialist care compared to primary care received Antabuse (n = 28 respectively n = 12, χ2 = 9.4, df = 1, P = 0.002). Fidelity In primary care, the general practitioners reported they fully followed the manual in 89.6 % (n = 120) of the first visits, partially in 6.7% (n = 9) and for 3.7 % (n = 5) data were missing. For the psychological treatment in primary care, 93.6% (n = 192) of the visits were rated as fully followed the manual, and 6.3% (n = 13) as partially. Primary outcome The ITT analysis, with available cases, showed that patients treated in primary care (n = 109) reduced their average weekly alcohol consumption from 367 to 224 g at 6 months follow up. The corresponding decrease in specialist care (n = 119) was from 343 to 182 g. Controlling for baseline consumption, the results on the pre-specified non-inferiority limit of 50 g for weekly alcohol consumption at 6-month follow up, is statistically inconclusive. This is due to the high end of the confidence interval exceeds 50 g (estimated mean weekly alcohol consumption was 30 g higher in primary care compared to specialist care, 95% CI: −10.20; 69.72). However, treatment in specialist care was not significantly superior to primary care (P = 0.146). Effect sizes For the primary outcome, the effect size based on available cases, in primary care was 0.66 and for specialist care 0.74. Sensitivity analysis Sensitivity analyses were performed in order to test the robustness of the results. The first analysis, ITT with multiple imputation by predictive mean matching, showed that patients treated in primary care (n = 144) had an estimated mean weekly alcohol consumption that was 27 g higher compared to patients treated in specialist care (n = 144) (95% CI: −10.90; 65.69, P = 0.159). The PP analysis showed that patients treated in primary care (n = 103) had an estimated mean weekly consumption that was 32 g higher compared to patients treated specialist care (n = 115) (95% CI: −7.21; 70.85), P = 0.110). The results were consistent with the ITT analysis, and are statistically inconclusive, since the confidence intervals exceed 50 g. Primary and secondary outcomes Table 2 shows the pre-specified primary and secondary outcome measures at baseline and at 6 months follow-up for available cases. Table 2. Outcome measures at baseline and 6 months follow up for primary care and specialist treatment. ITT data set with available cases Variable [mean (SD) range] (n = analysed)  Primary care, baseline  Primary care, 6 months follow up  Specialist care, baseline  Specialist care, 6 months follow up  Primary care, difference  Specialist care, difference  P-value  Weekly alcohol consumption (g)  367.4 (215.8) 47.6–1173.2 (n = 109)  224.2 (224.5) 0–1268.4 (n = 109)  343.3 (217.0) 0–1680 (n = 119)  181.9 (150.7) 0–772.8 (n = 119)  −143.3 (202.6) −772.8 to 680.4 (n = 109)  −161.3 (168.5) −907.2 to 165.2 (n = 119)  0.464  Heavy drinking days (per month)  13.1 (8.0) 1–30 (n = 109)  7.9 (8.9) 0–30 (n = 109)  11.9 (8.0) 0–30 (n = 119)  6.8 (7.8) 0–30 (n = 119)  −5.2 (8.9) −30 to 19 (n = 109)  −5.1 (7.8) −30 to 16 (n = 119)  0.954  ICD-10 criteria dependence  4.2 (1.0) 3–6 (n = 109)  2.1 (1.7) 0–6 (n = 105)  4.3 (1.0) 3–6 (n = 119)  2.2 (1.7) 0–6 (n = 118)  −2.1 (1.8) −6 to 2 (n = 105)  −2.1 (1.8) −6 to 3 (n = 118)  0.900  AUDIT total score  22.4 (5.8) 6–37 (n = 106)  14.1 (7.0) 0–32 (n = 100)  23.2 (4.9) 10–33 (n = 115)  13.5 (6.5) 0–35 (n = 111)  −8.4 (5.9) −30 to 3 (n = 99)  −9.3 (7.6) −30 to 7 (n = 107)  0.325  SADD total score  14.4 (6.3) 3–35 (n = 105)  9.0 (5.7) 0–34 (n = 100)  14.2 (5.3) 4–33 (n = 115)  8.0 (4.3) 0–23 (n = 114)  −5.5 (5.2) −20 to 3 (n = 98)  −6.2 (5.8) −23 to 5 (n = 110)  0.392  SIP total score  14.9 (7.0) 2–32 (n = 108)  8.0 (5.8) 0–29 (n = 103)  16.0 (6.5) 3–33 (n = 118)  7.8 (5.6) 0–22 (n = 115)  −6.8 (6.4) −26 to 7 (n = 103)  −7.9 (7.1) −30 to 7 (n = 114)  0.234  HADS scale anxiety  6.9 (4.1) 0–19 (n = 108)  5.7 (3.9) 0–19 (n = 104)  7.3 (4.4) 0–19 (n = 118)  5.6 (3.5) 0–17 (n = 114)  −1.1 (3.3) −9 to 15 (n = 103)  −1.4 (3.8) −14 to 7 (n = 113)  0.535  HADS scale depression  4.3 (3.4) 0–16 (n = 108)  3.1 (2.6) 0–10 (n = 105)  4.6 (3.2) 0–15 (n = 118)  3.1 (2.9) 0–15 (n = 116)  −1.3 (2.8) −9 to 8 (n = 104)  −1.5 (3.4) −12 to 7 (n = 115)  0.629  EQ 5D 5 L VAS  63.2 (18.8) 10–95 (n = 107)  73.7 (16.9) 20–100 (n = 105)  64.9 (18.4) 10–100 (n = 119)  75.5 (15.0) 30–100 (n = 116)  10.4 (19.5) −70 to 55 (n = 103)  10.5 (16.1) −35 to 65 (n = 116)  0.990  CDT  2.5 (2.1) 0.6–10.8 (n = 108)  2.4 (2.0) 0.6–12.5 (n = 99)  2.3 (2.0) 0.9–13.9 (n = 117)  1.9 (1.7) 0.6–13.6 (n = 107)  −0.2 (1.6) −6.9 to 4.6 (n = 98)  −0.28 (1.00) −6.2 to 2.4 (n = 105)  0.668  GT  0.8 (1.4) 0.1–11.8 (n = 109)  0.6 (0.9) 0.1–5.6 (n = 99)  0.8 (1.3) 0.1–8.6 (n = 118)  0.6 (1.0) 0.1–7.4 (n = 107)  −0.2 (1.4) −11.3 to 2.7 (n = 99)  −0.2 (0.6) −2.8 to 1.1 (n = 106)  0.710  ASAT  0.5 (0.3) 0.2–1.6 (n = 109)  0.5 (0.4) 0.2–3.6 (n = 98)  0.5 (0.3) 0.2–2.1 (n = 118)  0.5 (0.3) 0.2–1.7 (n = 107)  −0.0 (0.4) −1.0 to 3.1 (n = 98)  −0.1 (0.3) −1.7 to 0.8 (n = 106)  0.808  ALAT  0.6 (0.4) 0.1–2.2 (n = 109)  0.5 (0.4) 0.1–2.7 (n = 99)  0.5 (0.4) 0.1–2.8 (n = 118)  0.5 (0.4) 0.1–2.5 (n = 107)  −0.0 (0.4) −1.7 to 2.3 (n = 99)  0.0 (0.3) −1.9 to 1.6 (n = 106)  0.569  CSQ total score    21.6 (6.0) 9–32 (n = 95)    26.1 (4.8) 12–32 (n = 109)      <0.001*  Proportion below the Swedish national guidelines for hazardous consumptiona [n (%)]  6 (4.2) (n = 109)  19 (17.4) (n = 109)  4 (2.8) (n = 119)  23 (19.3) (n = 119)  −13 (−13.2) (n = 109)  −19 (−16.5)(n = 119)  0.843 df(1)  Variable [mean (SD) range] (n = analysed)  Primary care, baseline  Primary care, 6 months follow up  Specialist care, baseline  Specialist care, 6 months follow up  Primary care, difference  Specialist care, difference  P-value  Weekly alcohol consumption (g)  367.4 (215.8) 47.6–1173.2 (n = 109)  224.2 (224.5) 0–1268.4 (n = 109)  343.3 (217.0) 0–1680 (n = 119)  181.9 (150.7) 0–772.8 (n = 119)  −143.3 (202.6) −772.8 to 680.4 (n = 109)  −161.3 (168.5) −907.2 to 165.2 (n = 119)  0.464  Heavy drinking days (per month)  13.1 (8.0) 1–30 (n = 109)  7.9 (8.9) 0–30 (n = 109)  11.9 (8.0) 0–30 (n = 119)  6.8 (7.8) 0–30 (n = 119)  −5.2 (8.9) −30 to 19 (n = 109)  −5.1 (7.8) −30 to 16 (n = 119)  0.954  ICD-10 criteria dependence  4.2 (1.0) 3–6 (n = 109)  2.1 (1.7) 0–6 (n = 105)  4.3 (1.0) 3–6 (n = 119)  2.2 (1.7) 0–6 (n = 118)  −2.1 (1.8) −6 to 2 (n = 105)  −2.1 (1.8) −6 to 3 (n = 118)  0.900  AUDIT total score  22.4 (5.8) 6–37 (n = 106)  14.1 (7.0) 0–32 (n = 100)  23.2 (4.9) 10–33 (n = 115)  13.5 (6.5) 0–35 (n = 111)  −8.4 (5.9) −30 to 3 (n = 99)  −9.3 (7.6) −30 to 7 (n = 107)  0.325  SADD total score  14.4 (6.3) 3–35 (n = 105)  9.0 (5.7) 0–34 (n = 100)  14.2 (5.3) 4–33 (n = 115)  8.0 (4.3) 0–23 (n = 114)  −5.5 (5.2) −20 to 3 (n = 98)  −6.2 (5.8) −23 to 5 (n = 110)  0.392  SIP total score  14.9 (7.0) 2–32 (n = 108)  8.0 (5.8) 0–29 (n = 103)  16.0 (6.5) 3–33 (n = 118)  7.8 (5.6) 0–22 (n = 115)  −6.8 (6.4) −26 to 7 (n = 103)  −7.9 (7.1) −30 to 7 (n = 114)  0.234  HADS scale anxiety  6.9 (4.1) 0–19 (n = 108)  5.7 (3.9) 0–19 (n = 104)  7.3 (4.4) 0–19 (n = 118)  5.6 (3.5) 0–17 (n = 114)  −1.1 (3.3) −9 to 15 (n = 103)  −1.4 (3.8) −14 to 7 (n = 113)  0.535  HADS scale depression  4.3 (3.4) 0–16 (n = 108)  3.1 (2.6) 0–10 (n = 105)  4.6 (3.2) 0–15 (n = 118)  3.1 (2.9) 0–15 (n = 116)  −1.3 (2.8) −9 to 8 (n = 104)  −1.5 (3.4) −12 to 7 (n = 115)  0.629  EQ 5D 5 L VAS  63.2 (18.8) 10–95 (n = 107)  73.7 (16.9) 20–100 (n = 105)  64.9 (18.4) 10–100 (n = 119)  75.5 (15.0) 30–100 (n = 116)  10.4 (19.5) −70 to 55 (n = 103)  10.5 (16.1) −35 to 65 (n = 116)  0.990  CDT  2.5 (2.1) 0.6–10.8 (n = 108)  2.4 (2.0) 0.6–12.5 (n = 99)  2.3 (2.0) 0.9–13.9 (n = 117)  1.9 (1.7) 0.6–13.6 (n = 107)  −0.2 (1.6) −6.9 to 4.6 (n = 98)  −0.28 (1.00) −6.2 to 2.4 (n = 105)  0.668  GT  0.8 (1.4) 0.1–11.8 (n = 109)  0.6 (0.9) 0.1–5.6 (n = 99)  0.8 (1.3) 0.1–8.6 (n = 118)  0.6 (1.0) 0.1–7.4 (n = 107)  −0.2 (1.4) −11.3 to 2.7 (n = 99)  −0.2 (0.6) −2.8 to 1.1 (n = 106)  0.710  ASAT  0.5 (0.3) 0.2–1.6 (n = 109)  0.5 (0.4) 0.2–3.6 (n = 98)  0.5 (0.3) 0.2–2.1 (n = 118)  0.5 (0.3) 0.2–1.7 (n = 107)  −0.0 (0.4) −1.0 to 3.1 (n = 98)  −0.1 (0.3) −1.7 to 0.8 (n = 106)  0.808  ALAT  0.6 (0.4) 0.1–2.2 (n = 109)  0.5 (0.4) 0.1–2.7 (n = 99)  0.5 (0.4) 0.1–2.8 (n = 118)  0.5 (0.4) 0.1–2.5 (n = 107)  −0.0 (0.4) −1.7 to 2.3 (n = 99)  0.0 (0.3) −1.9 to 1.6 (n = 106)  0.569  CSQ total score    21.6 (6.0) 9–32 (n = 95)    26.1 (4.8) 12–32 (n = 109)      <0.001*  Proportion below the Swedish national guidelines for hazardous consumptiona [n (%)]  6 (4.2) (n = 109)  19 (17.4) (n = 109)  4 (2.8) (n = 119)  23 (19.3) (n = 119)  −13 (−13.2) (n = 109)  −19 (−16.5)(n = 119)  0.843 df(1)  aHDD = 0 and men ≤14 units/week; women ≤9 units/week (1 unit = 12 g alcohol. Table 2. Outcome measures at baseline and 6 months follow up for primary care and specialist treatment. ITT data set with available cases Variable [mean (SD) range] (n = analysed)  Primary care, baseline  Primary care, 6 months follow up  Specialist care, baseline  Specialist care, 6 months follow up  Primary care, difference  Specialist care, difference  P-value  Weekly alcohol consumption (g)  367.4 (215.8) 47.6–1173.2 (n = 109)  224.2 (224.5) 0–1268.4 (n = 109)  343.3 (217.0) 0–1680 (n = 119)  181.9 (150.7) 0–772.8 (n = 119)  −143.3 (202.6) −772.8 to 680.4 (n = 109)  −161.3 (168.5) −907.2 to 165.2 (n = 119)  0.464  Heavy drinking days (per month)  13.1 (8.0) 1–30 (n = 109)  7.9 (8.9) 0–30 (n = 109)  11.9 (8.0) 0–30 (n = 119)  6.8 (7.8) 0–30 (n = 119)  −5.2 (8.9) −30 to 19 (n = 109)  −5.1 (7.8) −30 to 16 (n = 119)  0.954  ICD-10 criteria dependence  4.2 (1.0) 3–6 (n = 109)  2.1 (1.7) 0–6 (n = 105)  4.3 (1.0) 3–6 (n = 119)  2.2 (1.7) 0–6 (n = 118)  −2.1 (1.8) −6 to 2 (n = 105)  −2.1 (1.8) −6 to 3 (n = 118)  0.900  AUDIT total score  22.4 (5.8) 6–37 (n = 106)  14.1 (7.0) 0–32 (n = 100)  23.2 (4.9) 10–33 (n = 115)  13.5 (6.5) 0–35 (n = 111)  −8.4 (5.9) −30 to 3 (n = 99)  −9.3 (7.6) −30 to 7 (n = 107)  0.325  SADD total score  14.4 (6.3) 3–35 (n = 105)  9.0 (5.7) 0–34 (n = 100)  14.2 (5.3) 4–33 (n = 115)  8.0 (4.3) 0–23 (n = 114)  −5.5 (5.2) −20 to 3 (n = 98)  −6.2 (5.8) −23 to 5 (n = 110)  0.392  SIP total score  14.9 (7.0) 2–32 (n = 108)  8.0 (5.8) 0–29 (n = 103)  16.0 (6.5) 3–33 (n = 118)  7.8 (5.6) 0–22 (n = 115)  −6.8 (6.4) −26 to 7 (n = 103)  −7.9 (7.1) −30 to 7 (n = 114)  0.234  HADS scale anxiety  6.9 (4.1) 0–19 (n = 108)  5.7 (3.9) 0–19 (n = 104)  7.3 (4.4) 0–19 (n = 118)  5.6 (3.5) 0–17 (n = 114)  −1.1 (3.3) −9 to 15 (n = 103)  −1.4 (3.8) −14 to 7 (n = 113)  0.535  HADS scale depression  4.3 (3.4) 0–16 (n = 108)  3.1 (2.6) 0–10 (n = 105)  4.6 (3.2) 0–15 (n = 118)  3.1 (2.9) 0–15 (n = 116)  −1.3 (2.8) −9 to 8 (n = 104)  −1.5 (3.4) −12 to 7 (n = 115)  0.629  EQ 5D 5 L VAS  63.2 (18.8) 10–95 (n = 107)  73.7 (16.9) 20–100 (n = 105)  64.9 (18.4) 10–100 (n = 119)  75.5 (15.0) 30–100 (n = 116)  10.4 (19.5) −70 to 55 (n = 103)  10.5 (16.1) −35 to 65 (n = 116)  0.990  CDT  2.5 (2.1) 0.6–10.8 (n = 108)  2.4 (2.0) 0.6–12.5 (n = 99)  2.3 (2.0) 0.9–13.9 (n = 117)  1.9 (1.7) 0.6–13.6 (n = 107)  −0.2 (1.6) −6.9 to 4.6 (n = 98)  −0.28 (1.00) −6.2 to 2.4 (n = 105)  0.668  GT  0.8 (1.4) 0.1–11.8 (n = 109)  0.6 (0.9) 0.1–5.6 (n = 99)  0.8 (1.3) 0.1–8.6 (n = 118)  0.6 (1.0) 0.1–7.4 (n = 107)  −0.2 (1.4) −11.3 to 2.7 (n = 99)  −0.2 (0.6) −2.8 to 1.1 (n = 106)  0.710  ASAT  0.5 (0.3) 0.2–1.6 (n = 109)  0.5 (0.4) 0.2–3.6 (n = 98)  0.5 (0.3) 0.2–2.1 (n = 118)  0.5 (0.3) 0.2–1.7 (n = 107)  −0.0 (0.4) −1.0 to 3.1 (n = 98)  −0.1 (0.3) −1.7 to 0.8 (n = 106)  0.808  ALAT  0.6 (0.4) 0.1–2.2 (n = 109)  0.5 (0.4) 0.1–2.7 (n = 99)  0.5 (0.4) 0.1–2.8 (n = 118)  0.5 (0.4) 0.1–2.5 (n = 107)  −0.0 (0.4) −1.7 to 2.3 (n = 99)  0.0 (0.3) −1.9 to 1.6 (n = 106)  0.569  CSQ total score    21.6 (6.0) 9–32 (n = 95)    26.1 (4.8) 12–32 (n = 109)      <0.001*  Proportion below the Swedish national guidelines for hazardous consumptiona [n (%)]  6 (4.2) (n = 109)  19 (17.4) (n = 109)  4 (2.8) (n = 119)  23 (19.3) (n = 119)  −13 (−13.2) (n = 109)  −19 (−16.5)(n = 119)  0.843 df(1)  Variable [mean (SD) range] (n = analysed)  Primary care, baseline  Primary care, 6 months follow up  Specialist care, baseline  Specialist care, 6 months follow up  Primary care, difference  Specialist care, difference  P-value  Weekly alcohol consumption (g)  367.4 (215.8) 47.6–1173.2 (n = 109)  224.2 (224.5) 0–1268.4 (n = 109)  343.3 (217.0) 0–1680 (n = 119)  181.9 (150.7) 0–772.8 (n = 119)  −143.3 (202.6) −772.8 to 680.4 (n = 109)  −161.3 (168.5) −907.2 to 165.2 (n = 119)  0.464  Heavy drinking days (per month)  13.1 (8.0) 1–30 (n = 109)  7.9 (8.9) 0–30 (n = 109)  11.9 (8.0) 0–30 (n = 119)  6.8 (7.8) 0–30 (n = 119)  −5.2 (8.9) −30 to 19 (n = 109)  −5.1 (7.8) −30 to 16 (n = 119)  0.954  ICD-10 criteria dependence  4.2 (1.0) 3–6 (n = 109)  2.1 (1.7) 0–6 (n = 105)  4.3 (1.0) 3–6 (n = 119)  2.2 (1.7) 0–6 (n = 118)  −2.1 (1.8) −6 to 2 (n = 105)  −2.1 (1.8) −6 to 3 (n = 118)  0.900  AUDIT total score  22.4 (5.8) 6–37 (n = 106)  14.1 (7.0) 0–32 (n = 100)  23.2 (4.9) 10–33 (n = 115)  13.5 (6.5) 0–35 (n = 111)  −8.4 (5.9) −30 to 3 (n = 99)  −9.3 (7.6) −30 to 7 (n = 107)  0.325  SADD total score  14.4 (6.3) 3–35 (n = 105)  9.0 (5.7) 0–34 (n = 100)  14.2 (5.3) 4–33 (n = 115)  8.0 (4.3) 0–23 (n = 114)  −5.5 (5.2) −20 to 3 (n = 98)  −6.2 (5.8) −23 to 5 (n = 110)  0.392  SIP total score  14.9 (7.0) 2–32 (n = 108)  8.0 (5.8) 0–29 (n = 103)  16.0 (6.5) 3–33 (n = 118)  7.8 (5.6) 0–22 (n = 115)  −6.8 (6.4) −26 to 7 (n = 103)  −7.9 (7.1) −30 to 7 (n = 114)  0.234  HADS scale anxiety  6.9 (4.1) 0–19 (n = 108)  5.7 (3.9) 0–19 (n = 104)  7.3 (4.4) 0–19 (n = 118)  5.6 (3.5) 0–17 (n = 114)  −1.1 (3.3) −9 to 15 (n = 103)  −1.4 (3.8) −14 to 7 (n = 113)  0.535  HADS scale depression  4.3 (3.4) 0–16 (n = 108)  3.1 (2.6) 0–10 (n = 105)  4.6 (3.2) 0–15 (n = 118)  3.1 (2.9) 0–15 (n = 116)  −1.3 (2.8) −9 to 8 (n = 104)  −1.5 (3.4) −12 to 7 (n = 115)  0.629  EQ 5D 5 L VAS  63.2 (18.8) 10–95 (n = 107)  73.7 (16.9) 20–100 (n = 105)  64.9 (18.4) 10–100 (n = 119)  75.5 (15.0) 30–100 (n = 116)  10.4 (19.5) −70 to 55 (n = 103)  10.5 (16.1) −35 to 65 (n = 116)  0.990  CDT  2.5 (2.1) 0.6–10.8 (n = 108)  2.4 (2.0) 0.6–12.5 (n = 99)  2.3 (2.0) 0.9–13.9 (n = 117)  1.9 (1.7) 0.6–13.6 (n = 107)  −0.2 (1.6) −6.9 to 4.6 (n = 98)  −0.28 (1.00) −6.2 to 2.4 (n = 105)  0.668  GT  0.8 (1.4) 0.1–11.8 (n = 109)  0.6 (0.9) 0.1–5.6 (n = 99)  0.8 (1.3) 0.1–8.6 (n = 118)  0.6 (1.0) 0.1–7.4 (n = 107)  −0.2 (1.4) −11.3 to 2.7 (n = 99)  −0.2 (0.6) −2.8 to 1.1 (n = 106)  0.710  ASAT  0.5 (0.3) 0.2–1.6 (n = 109)  0.5 (0.4) 0.2–3.6 (n = 98)  0.5 (0.3) 0.2–2.1 (n = 118)  0.5 (0.3) 0.2–1.7 (n = 107)  −0.0 (0.4) −1.0 to 3.1 (n = 98)  −0.1 (0.3) −1.7 to 0.8 (n = 106)  0.808  ALAT  0.6 (0.4) 0.1–2.2 (n = 109)  0.5 (0.4) 0.1–2.7 (n = 99)  0.5 (0.4) 0.1–2.8 (n = 118)  0.5 (0.4) 0.1–2.5 (n = 107)  −0.0 (0.4) −1.7 to 2.3 (n = 99)  0.0 (0.3) −1.9 to 1.6 (n = 106)  0.569  CSQ total score    21.6 (6.0) 9–32 (n = 95)    26.1 (4.8) 12–32 (n = 109)      <0.001*  Proportion below the Swedish national guidelines for hazardous consumptiona [n (%)]  6 (4.2) (n = 109)  19 (17.4) (n = 109)  4 (2.8) (n = 119)  23 (19.3) (n = 119)  −13 (−13.2) (n = 109)  −19 (−16.5)(n = 119)  0.843 df(1)  aHDD = 0 and men ≤14 units/week; women ≤9 units/week (1 unit = 12 g alcohol. The results from the secondary outcomes follow the same pattern as the primary outcome, with no statistically significant differences between the groups. The only exception was client satisfaction, with significant differences in CSQ scores, where patients treated in specialist care were more satisfied with the treatment (t = 5.9, P < 0.001). Predictors for treatment outcome Additional analyses were performed for available cases, investigating whether length of treatment and compliance with treatment affected the primary outcome. As these are results from sub-analyses, caution is warranted in the interpretations. In both treatment conditions a higher number of visits predicted lower weekly alcohol consumption at 6 months (n = 228, −12.69, 95% CI: −21.76; −3.63, P = 0.006). However, length of treatment measured in minutes did not predict the primary outcome (n = 228, −0.16, 95% CI: −0.37; 0.06 P = 0.146). When number of visits were controlled for, the results confirm that primary care was non-inferior to specialist care (n = 228, 4.73, 95% CI: −38.71; 48.18, P = 0.830). Non-inferiority was also confirmed for participants who self rated that they had completed the treatment they were randomized to (n = 144, −5.43, 95% CI: −42.28; 31.42, P = 0.771). Analyses were also performed to investigate whether participants’ gender, age, severity of dependence, level of anxiety or depression and readiness to change, affected the primary outcome weekly alcohol consumption (grams of alcohol) in primary care compared to specialist care (Fig. 2). Fig. 2. View largeDownload slide Sub-analyses of predictors for differences in primary outcome in primary care compared to specialist treatment. ITT data set with available cases. The value on the coefficient indicates the estimated mean difference between primary care and specialist care. Fig. 2. View largeDownload slide Sub-analyses of predictors for differences in primary outcome in primary care compared to specialist treatment. ITT data set with available cases. The value on the coefficient indicates the estimated mean difference between primary care and specialist care. Participants with AUDIT scores indicative of a low severity of dependence (<20), showed a significantly larger reduction of weekly alcohol consumption in primary care compared to specialist care. Women and participants with low severity of dependence irrespective of gender, measured with SADD, showed non-inferior outcomes in primary care compared with specialist care, as the high ends of the confidence intervals were below 50 g of alcohol. On the other hand, patients with a high severity of alcohol dependence, measured with AUDIT or SADD, or high levels of depression showed a significantly larger reduction in alcohol consumption when treated in specialist care. There was no difference in treatment outcome in primary care compared to specialist care for younger participants, compared to middle aged or older. Regarding readiness to change, a higher rating of confidence predicted a lower weekly alcohol consumption at 6 months (n = 219, −1.38, 95% CI: −2.40; −0.37, P = 0.007). Importance did not predict alcohol consumption at 6 months (n = 220, −0.75, 95% CI: −2.80; 1.30, P = 0.474). The readiness to change measures did not alter the estimate for the difference between primary care and specialist care on the primary outcome measure (n = 219, 27.63, 95% CI: −12.94; 68.19, P = 0.181) Lost to follow up The follow-up rate was 81% overall. In the primary care condition 24% (n = 34) were lost to follow up and 15% (n = 22) in specialist care. However, this is not statistically significant (χ2 = 1.705, df = 1, P = 0.19). Participants randomized to primary care who were lost to follow up had significantly lower weekly alcohol consumption at baseline compared to those who completed followed up (mean 269 respectively 367, t = 2.49, P = 0.014). They also had a lower consumption compared to participants lost to follow up after specialist care (mean 377, t = 2.25, P = 0.028). In both treatment conditions, participants lost to follow up were younger (age mean 50 respectively 56, t = 4.00, P = 0.000), had higher levels of depression at baseline (HADS depression mean 6.1 respectively 4.8, t = 3.42, P = 0.001), and higher levels of anxiety at baseline (HADS anxiety mean 9.0 respectively 7.1, t = 3.00, P = 0.003), compared to those who completed followed up. Side effects At follow up, seven participants from primary care and six from specialist care, reported side effects from the pharmacological treatment (one could report more than one side effect): nausea (n = 4); diarrhoea (n = 2); anxiety (n = 2); thrombocytopenia (n = 1); perspirations (n = 1); shaky (n = 1); stomach pain (n = 1); sleeping difficulties (n = 1); tiredness (n = 1); and vertigo (n = 1). Six participants did not specify which type of side effects. One patient, who was randomized to primary care, died during the study period (cause unknown). DISCUSSION Overall findings The aim of this study was to evaluate if treatment of alcohol dependence in primary care, delivered by general practitioners, is equally effective as treatment by specialists in reducing alcohol consumption. The results for the primary outcome were statistically inconclusive, where we cannot conclude that primary care treatment is non-inferior to specialist care. On the other hand, treatment in specialist care was not found to be superior to treatment in primary care. Only on satisfaction with treatment, was specialist care superior to primary care. However, the results from sub-analysis are very informative. Primary care treatment was non-inferior to specialist care among participants who reported that they completed the treatment. The general practitioners only received 1 day of training, and no further supervision during the trial. It is possible that more training or regular supervision could have improved the general practitioners' skills, and thereby improved retention to treatment for the participants. This study also illustrates the role of severity of dependence for treatment outcome in different settings. Sub-analyses showed that individuals with alcohol dependence of low severity reduced their consumption to the same extent in primary care as in specialist care. In contrast, individuals with alcohol dependence of high severity showed better outcomes after specialist treatment. Also, patients with higher level of depression showed better treatment outcomes in specialist care. Previous studies of treatment in primary care have shown varied results (Oslin et al., 2013; O’Malley et al., 2003; Drummond et al., 1990). The manual in this study is briefer compared to other manuals for treatment of alcohol dependence in primary care. In O’Malley et al. (2003) the primary care treatment was an initial 45-min visit followed by seven 15–20 min sessions. While in Oslin et al. (2013) the primary care treatment covered on average ten 30 min visits. As number of visits contributed significantly to the outcome, it is possible the 15-method was too brief to facilitate change among individuals with more severe dependence and higher levels of depression. On the other hand, it appeared sufficient for the larger group with moderate levels of dependence. This suggests that the more severe group should be offered additional treatment or be referred to specialized care. This study included individuals with alcohol dependence who traditionally have been reluctant to seek treatment. Participants showed lower weekly alcohol consumption, fewer psychiatric co-morbidities, higher education and higher rates of co-habiting compared to previous studies on treatment of alcohol dependence in primary care (Oslin et al., 2013; Coste et al., 2016). At baseline, the weekly alcohol consumption in this sample corresponded to the level found among primary care patients with alcohol dependence who are not treatment seekers for alcohol problems (Rehm et al., 2015b). This is possibly because the main recruitment method was through advertisements, which is known to attract individuals with lower problem severity compared to clinical samples (Morley et al., 2009). While the majority of participants were still drinking at hazardous levels at 6 months follow-up, they had reduced their consumption substantially. From a public health perspective, it needs to be stressed that heavy drinking is a strong risk factor for mortality, and that any decrease in consumption is associated with a reduced risk for mortality (Laramee et al., 2015). Limitations This study reports outcomes of primary care treatment for identified alcohol dependent patients. A major challenge in primary care however, is the identification of alcohol problems, which is the first step of the 15-method, but not part of this study. As long as practitioners are unable to identify patients with alcohol problems, having access to effective treatment will only reach individuals who are concerned about their drinking and actively seek primary care treatment. A limitation in the study design lies in the choice of the comparison treatment. The specialized addiction unit was recently developed to be an attractive alternative to treatment as usual in specialized addiction care, for the same segment of the dependence population as the sample of this study. This may have improved the retention to treatment and thereby the outcomes in the specialist treatment condition, compared to if the participants had been randomized to regular specialist care. In order to keep the study as naturalistic as possible, the data collected on treatment fidelity were only self-reported. However, this is also a limitation to the study, as there are no objective ratings of fidelity to the manual. As the participants were alcohol dependent, and consumed alcohol at harmful levels, the design did not, due to ethical reasons, include a no-treatment control. Even though natural recovery is common among individuals with alcohol dependence, seeking treatment improves the rate of recovery, independent of severity of dependence (Dawson et al., 2006). In Dawson et al. (2006), 45.7% of treatment seekers recovered, while the corresponding numbers for non-treatment seekers were 32.5%. It is therefore important to reach people with alcohol dependence with treatments that they find attractive and are willing to seek. Lastly, nearly 20% of the participants were lost to follow up. This level of drop out is correspondent to previous studies, for example UKATT (UKATT, 2005), but nevertheless constitutes an impediment to the conclusions. Generalizability The general practitioners who delivered the interventions were ordinary clinicians, who received 1-day training in the method. The generalizability to clinical settings is therefore very good. The participants had lower severity of problems compared to other studies. From an epidemiological perspective, this represents the majority of individuals with alcohol dependence (Andreasson et al., 2013). However, as the participants’ actively sought treatment by responding to an advertisement, and showed high scores on readiness to change, the results may not be generalized to proactively recruited individuals. Interpretation Primary care practitioners can, compared to addiction specialists, successfully treat individuals with a low to moderate severity of alcohol dependence. The 15-method, may be a way to broaden the base of treatment for alcohol dependence, reducing the treatment gap in this field. By offering services in primary care, it is possible to reach a larger share of all individuals with alcohol dependence. This is especially important in jurisdictions with little or no specialized addiction treatment. ACKNOWLEDGEMENTS We want to thank the participating primary care units and the general practitioners who delivered the treatments. We also want to thank Victoria Andersson, research nurse, who was the study coordinator. CONFLICT OF INTEREST STATEMENT None declared. OTHER INFORMATION The study protocol is available from the corresponding author. REFERENCES Alvarez-Bueno C, Rodriguez-Martin B, Garcia-Ortiz L, et al.  . ( 2015) Effectiveness of brief interventions in primary health care settings to decrease alcohol consumption by adult non-dependent drinkers: a systematic review of systematic reviews. Prev Med  76: S33– 8. Google Scholar CrossRef Search ADS PubMed  Andreasson S, Danielsson AK, Hallgren M. ( 2013) Severity of alcohol dependence in the Swedish adult population: association with consumption and social factors. Alcohol  47: 21– 5. Google Scholar CrossRef Search ADS PubMed  Andreasson S, Hansagi H, Osterlund B. ( 2002) Short-term treatment for alcohol-related problems: four-session guided self-change versus one session of advice—a randomized, controlled trial. Alcohol  28: 57– 62. Google Scholar CrossRef Search ADS PubMed  Babor TF, Higgins-Biddle JC, Saunders JB, et al.  . ( 2001) AUDIT—The Alcohol Use Disorders Identification Test. Guidelines for use in Primary Care . Switzerland: World Health Organization. Berger L, Fisher M, Brondino M, et al.  . ( 2013) Efficacy of acamprosate for alcohol dependence in a family medicine setting in the United States: a randomized, double-blind, placebo-controlled study. Alcohol Clin Exp Res  37: 668– 74. Google Scholar CrossRef Search ADS PubMed  Bischof G, Grothues JM, Reinhardt S, et al.  . ( 2008) Evaluation of a telephone-based stepped care intervention for alcohol-related disorders: a randomized controlled trial. Drug Alcohol Depend  93: 244– 51. Google Scholar CrossRef Search ADS PubMed  Chick J, Ritson B, Connaughton J, et al.  . ( 1988) Advice versus extended treatment for alcoholism: a controlled study. Br J Addict  83: 159– 70. Google Scholar CrossRef Search ADS PubMed  Coste F, Rahhali N, Wallace P. ( 2016) STREAM: elucidating how alcohol use disorders are managed in UK General Practice Today. Alcohol Alcohol . 51: 428– 35. Google Scholar CrossRef Search ADS PubMed  Dawson DA, Grant BF, Stinson FS, et al.  . ( 2006) Estimating the effect of help-seeking on achieving recovery from alcohol dependence. Addiction  101: 824– 34. Google Scholar CrossRef Search ADS PubMed  Drummond DC, Thom B, Brown C, et al.  . ( 1990) Specialist versus general practitioner treatment of problem drinkers. Lancet  336: 915– 8. Google Scholar CrossRef Search ADS PubMed  Glass JE, Hamilton AM, Powell BJ, et al.  . ( 2015) Specialty substance use disorder services following brief alcohol intervention: a meta-analysis of randomized controlled trials. Addiction  110: 1404– 15. Google Scholar CrossRef Search ADS PubMed  Herdman M, Gudex C, Lloyd A, et al.  . ( 2011) Development and preliminary testing of the new five-level version of EQ-5D (EQ-5D-5L). Qual Life Res  20: 1727– 36. Google Scholar CrossRef Search ADS PubMed  Karhuvaara S, Simojoki K, Virta A, et al.  . ( 2007) Targeted nalmefene with simple medical management in the treatment of heavy drinkers: a randomized double-blind placebo-controlled multicenter study. Alcohol Clin Exp Res  31: 1179– 87. Google Scholar CrossRef Search ADS PubMed  Kiritze-Topor P, Huas D, Rosenzweig C, et al.  . ( 2004) A pragmatic trial of acamprosate in the treatment of alcohol dependence in primary care. Alcohol Alcohol  39: 520– 7. Google Scholar CrossRef Search ADS PubMed  Kohn R, Saxena S, Levav I, et al.  . ( 2004) The treatment gap in mental health care. Bull World Health Organ  82: 858– 66. Google Scholar PubMed  Laramee P, Leonard S, Buchanan-Hughes A, et al.  . ( 2015) Risk of all-cause mortality in alcohol-dependent individuals: a systematic literature review and meta-analysis. EBioMedicine  2: 1394– 404. Google Scholar CrossRef Search ADS PubMed  Larsen DL, Attkisson CC, Hargreaves WA, et al.  . ( 1979) Assessment of client/patient satisfaction: development of a general scale. Eval Program Plann  2: 197– 207. Google Scholar CrossRef Search ADS PubMed  Lieberman DZ, Cioletti A, Massey SH, et al.  . ( 2014) Treatment preferences among problem drinkers in primary care. Int J Psychiatry Med  47: 231– 40. Google Scholar CrossRef Search ADS PubMed  Magill M, Ray LA. ( 2009) Cognitive-behavioral treatment with adult alcohol and illicit drug users: a meta-analysis of randomized controlled trials. J Stud Alcohol Drugs  70: 516– 27. Google Scholar CrossRef Search ADS PubMed  Miller WR, Sovereign RG, Krege B. ( 1988) Motivational interviewing with problem drinkers: II. The Drinker’s Check-up as a preventive intervention. Behavioural Psychotherapy  16: 251– 68. Google Scholar CrossRef Search ADS   Miller WR, Tonigan JS, Longabaugh R. ( 1995) The Drinker Inventory of Consequences (DrInC): an instrument for assessing adverse consequences of alcohol abuse (Test Manual). NIAA Project MATCH Monograph Series . Bethesda: : Department of Health and Human Service. Morley KC, Teesson M, Sannibale C, et al.  . ( 2009) Sample bias from different recruitment strategies in a randomised controlled trial for alcohol dependence. Drug Alcohol Rev  28: 222– 9. Google Scholar CrossRef Search ADS PubMed  Oslin DW, Lynch KG, Maisto SA, et al.  . ( 2013) A randomized clinical trial of alcohol care management delivered in Department of Veterans Affairs Primary Care Clinics Versus Specialty Addiction Treatment. J Gen Intern Med  29: 162– 8. Google Scholar CrossRef Search ADS PubMed  O’Donnell A, Anderson P, Newbury-Birch D, et al.  . ( 2014) The impact of brief alcohol interventions in primary healthcare: a systematic review of reviews. Alcohol Alcohol  49: 66– 78. Google Scholar CrossRef Search ADS PubMed  O’Malley SS, Rounsaville BJ, Farren C, et al.  . ( 2003) Initial and maintenance naltrexone treatment for alcohol dependence using primary care vs specialty care: a nested sequence of 3 randomized trials. Arch Intern Med  163: 1695– 704. Google Scholar CrossRef Search ADS PubMed  Probst C, Manthey J, Martinez A, et al.  . ( 2015) Alcohol use disorder severity and reported reasons not to seek treatment: a cross-sectional study in European primary care practices. Subst Abuse Treat Prev Policy  10: 32. Google Scholar CrossRef Search ADS PubMed  Projectmatch. ( 1997) Matching alcoholism treatments to client heterogeneity: project MATCH posttreatment drinking outcomes. J Stud Alcohol  58: 7– 29. CrossRef Search ADS PubMed  Raistrick DD, Davidson GR. ( 1983) Development of a questionnaire to measure alcohol dependence. Br J Addict  78: 89– 95. Google Scholar CrossRef Search ADS PubMed  Rehm J, Allamani A, Elekes Z, et al.  . ( 2015a) Alcohol dependence and treatment utilization in Europe—a representative cross-sectional study in primary care. BMC Fam Pract  16: 90. Google Scholar CrossRef Search ADS PubMed  Rehm J, Manthey J, Struzzo P, et al.  . ( 2015b) Who receives treatment for alcohol use disorders in the European Union? A cross-sectional representative study in primary and specialized health care. Eur Psychiatry  30: 885– 93. Google Scholar CrossRef Search ADS PubMed  Rehm J, Shield KD, Gmel G, et al.  . ( 2013) Modeling the impact of alcohol dependence on mortality burden and the effect of available treatment interventions in the European Union. Eur Neuropsychopharmacol  23: 89– 97. Google Scholar CrossRef Search ADS PubMed  Riddargatan 1 ( 2012). Manual för 15-metoden (Swedish). http://www.ridddargatan1.se/PV/: Riddargatan 1, Stockholm Centre for Dependency Disorders/Centre for Psychiatry Research, Stockholm. Saitz R. ( 2010) Alcohol screening and brief intervention in primary care: absence of evidence for efficacy in people with dependence or very heavy drinking. Drug Alcohol Rev  29: 631– 40. Google Scholar CrossRef Search ADS PubMed  Saitz R, Cheng DM, Winter M, et al.  . ( 2013) Chronic care management for dependence on alcohol and other drugs: the AHEAD randomized trial. J Am Med Assoc  310: 1156– 67. Google Scholar CrossRef Search ADS   Sellman JD, Sullivan PF, Dore GM, et al.  . ( 2001) A randomized controlled trial of motivational enhancement therapy (MET) for mild to moderate alcohol dependence. J Stud Alcohol  62: 389– 96. Google Scholar CrossRef Search ADS PubMed  Sobell MB, Sobell LC. ( 1993) Problem Drinkers: Guided Self-Change Treatment . New York: Guilford. Sobell LC, Sobell MB 1996. Time Line Follow Back. User´s Guide, Toronto, Addiction Research Foundation. Storbjork J, Room R. ( 2008) The two world of alcohol problems: Who is in treatment and who is not? Addiction Research & Theory  16: 67– 84. Google Scholar CrossRef Search ADS   The World Health Organization. ( 1992) The ICD-10 Classification of Mental and Behavioural Disorders . Geneva: World Health Organization. Trim RS, Schuckit MA, Smith TL. ( 2013) Predictors of initial and sustained remission from alcohol use disorders: findings from the 30-year follow-up of the San Diego Prospective Study. Alcohol Clin Exp Res  37: 1424– 31. Google Scholar CrossRef Search ADS PubMed  Ukatt. ( 2005) Effectiveness of treatment for alcohol problems: findings of the randomised UK alcohol treatment trial (UKATT). Br Med J  331: 541. CrossRef Search ADS   Upshur C, Weinreb L, Bharel M, et al.  . ( 2015) A randomized control trial of a chronic care intervention for homeless women with alcohol use problems. J Subst Abuse Treat  51: 19– 29. Google Scholar CrossRef Search ADS PubMed  Wallhed Finn S, Bakshi A-S, Andréasson S ( 2014). Alcohol consumption, dependence, and treatment barriers: perceptions among nontreatment seekers with alcohol dependence. Substance Use & Misuse 49: 762– 9. Walters GD. ( 2000) Behavioral self-control training for problem drinkers: a meta-analysis of randomized control studies. Behav Ther  31: 135– 49. Google Scholar CrossRef Search ADS   Watkins KE, Ober AJ, Lamp K, et al.  . ( 2017) Collaborative care for opioid and alcohol use disorders in primary care: The SUMMIT Randomized Clinical Trial. JAMA Intern Med  177: 1480– 88. Google Scholar CrossRef Search ADS PubMed  Willenbring ML, Olson DH. ( 1999) A randomized trial of integrated outpatient treatment for medically ill alcoholic men. Arch Intern Med  159: 1946– 52. Google Scholar CrossRef Search ADS PubMed  Witkiewitz K, Falk DE, Kranzler HR, et al.  . ( 2014) Methods to analyze treatment effects in the presence of missing data for a continuous heavy drinking outcome measure when participants drop out from treatment in alcohol clinical trials. Alcohol Clin Exp Res  38: 2826– 34. Google Scholar CrossRef Search ADS PubMed  Zigmond AS, Snaith RP. ( 1983) The hospital anxiety and depression scale. Acta Psychiatr Scand  67: 361– 70. Google Scholar CrossRef Search ADS PubMed  Author notes Clinical Trial Registration: Trial identifier ISRCTN84490505 at Controlled Trials.com, assigned 2013-07-09. © The Author(s) 2018. Medical Council on Alcohol and Oxford University Press. All rights reserved.

Journal

Alcohol and AlcoholismOxford University Press

Published: Jan 16, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off