915
CLINICAL PRACTICE
Specialist versus general practitioner treatment of problem drinkers
efficacy of specialist versus general practitioner (GP) treatment of problem drinkers
The
assessed in a randomised controlled trial. 40 problem drinkers referred consecutively to a specialist alcohol clinic by their GP were, after assessment, randomly allocated to either GP or specialist clinic treatment groups. All subjects received intitial advice and counselling in the clinic about their drinking. The specialist clinic group received continued care from the clinic including, if necessary, admission to hospital. Patients in the GP group were returned to the care of the G P who was contacted and supported by the specialist. After 6 months of follow-up, there were significant reductions in alcohol consumption and alcohol-related problems in both groups. No significant difference was found between the two groups with respect to the main outcome measures. No differential treatment effect was found with the more severely dependent drinkers. The findings show that after an initial detailed assessment and advice session, the treatment provided by GPs is at least as effective as that from a specialist clinic with respect to improvements in drinking behaviour and alcohol-related problems. After initial assessment and advice, specialist clinics should encourage GPs to become more involved in the subsequent care of problem drinkers. Such a practice should be based on the individual patient’s needs and the adequacy of support offered to GPs. was
Introduction The role of the general practitioner (GP) in the identification and treatment of problem drinkers has been the focus of much interest and debate.l-6 The Kessel report’ suggested
that GPs were ideally placed to provide treatment and care for this population and gave impetus to the development of community-based services for problem drinkers in the 1980s. However, GPs seem to be reluctant to take on this extra burden.’-1 Lack of resources and support, inadequate education, and negative views about treatment efficacy have
been blamed for this disinclination. The concept of the Community Alcohol Team has been criticised," but the specialist support and training which it provides has been shown to have a beneficial effect on GPs’ therapeutic commitment to problem drinkers.12 Although advice and information given to heavy drinkers by GPs reduces alcohol consumption,13 is the treatment provided by GPs as effective as that from a specialist clinic? This important question has not previously been studied in a controlled trial. Furthermore, it is not known whether patients benefit from specialist or GP treatment differentially according to the extent of dependence or problem severity. Such a finding could have implications for appropriate treatment
matching. The study reported here was conducted as part of a larger investigation of the attitudes and behaviour of the providers and recipients of GP and specialist treatment of drinking problems (unpublished). The null hypothesis of this controlled trial was that GP and specialist clinic care were equally effective. Orford and colleagues14 found that severely dependent drinkers given intensive specialist treatment had a better outcome than did those given simple advice. A further hypothesis of our trial, therefore, was that severely dependent subjects would benefit more from specialist clinic treatment. Patients and methods Patients 40 problem drinkers, referred consecutively to the Maudsley Hospital Alcohol Clinic by their GPs, were randomly assigned to one of four specialists (one clinical nurse specialist and three psychiatrists). All patients received a full clinical assessment, including psychiatric and physical examination, and relevant blood investigations. Also, patients were asked to complete the Severity of Alcohol Dependence Questionnaire (SADQ), the General Health Questionnaire (GHQ), and the Alcohol Problems Questionnaire ADDRESSES Addiction Research Unit, National Addiction Centre, Institute of Psychiatry, and Maudsley Hospital, Denmark Hill, London, UK (D. C. Drummond, MRCPsych, B. Thom, MA, C Brown, RMN, Prof G Edwards, FRCPsych, M. J. Mullan, MB) Correspondence to Dr D C Drummond Addiction Research Unit, National Addiction Centre, Institute of Psychiatry, 101 Denmark Hill, London SE5 8AF, UK
916
(APQ). Informed consent was obtained and patients were randomly allocated to one of two treatment groups-namely, the clinic group and the GP group. Subsets of patients defined by a SADQ score (two levels) were separately randomised to maximise the likelihood of group matching. The clinic group received counselling and advice about their drinking problem followed by routine outpatient clinic care and, if necessary, admission to hospital. The GP group were also counselled and were given advice about their drinking problem, but after this initial consultation, they were returned to the care of their GP. These patients were assured that the specialist would remain involved in their treatment via contact with their GP and they were told to contact their GP for an appointment. Specialists then arranged to visit the GP to discuss the management of the patient, and they provided a short booklet of guidelines on the management of alcohol problems. The specialist offered further advice and support to the GP, initiated further contact by telephone to check on progress or difficulties, and provided a contact telephone number. Although patients in the GP group were encouraged to attend the GP rather than the clinic, a minority decided to attend the Maudsley and other specialist clinics during
follow-up. The initial assessment interview, including counselling, took a of 3 h. All patients and GPs were told that they would be contacted after 6 months to assess outcome. Follow-up interviews with patients were conducted by two research workers who were not blinded to the patient’s allocated group but who were not connected with the patient’s care. This interview sought information about alcohol consumption and related problems, treatment contact, and perceptions of the care offered. Patients were also asked to complete the APQ, GHQ, and a questionnaire designed to measure their perception of change in problem status. mean
Measures The SADQ15 consists of a 20-item self-completion questionnaire which measures the frequency and severity of symptoms of the Alcohol Dependence Syndrome’6 during a recent period of heavy drinking. The GHQ (see Goldberg17) has been widely used in studies of the prevalence of psychiatric disorder in clinical populations. The APQ is a new questionnaire which has been TABLE I-DEMOGRAPHIC DATA AND DRINKING HISTORY AT INTAKE
alcohol-related problems of the previousf: consists of 46 items about a wide range of common!;’ months;18 it affected problem areas, although only an aggregate score of the 23 common items which were applicable to all subjects are reported here (APQC score). Alcohol consumption during the 6 months before and after the initial consultation was measured by a standardised interview method. The patient was asked to recall the maximum amount of alcohol consumed on any one day during each of the 26 week Patients were also asked to rate change in problem status at follow-up as either ’improved’, ’worse’, or ’unchanged’ on a 13-item, 3-point Lickert scale. An overall self-rated change score is the unweighted sum of the 13 variables.
designed
to measure
Results
Adequacy of matching at intake (table i) The two groups were broadly similar with respect characteristics (Student’s t-test). to various intake Nevertheless, because there was a difference between the groups in mean age and time in present accommodation, these variables were used as covariates in subsequent analyses. Since SADQ score was a matching variable, this was also used as a covariate in subsequent analyses. Follow-up Outcome. Follow-up data were obtained for 19 of the clinic patients and 18 of the GP patients. 1 patient in the clinic group and 1 in the GP group refused a follow-up interview, although information about the drinking behaviour of the patient in the GP group during follow-up was obtained from the GP. 2 GP patients could not be traced and no relevant information was available from their GPs. GHQ and APQ data were obtained for 17 and 16 cases, respectively from each group at follow-up. Since it proved impossible to interview all subjects at the 6-month followup point, the mean length of time (SEM) after initial assessment when follow-up interviews were conducted was significantly longer in the clinic group than in the GP group (10-0 months [0’8] vs 7-7 [0’5]; p
CLINICAL PRACTICE
Specialist versus general practitioner treatment of problem drinkers
efficacy of specialist versus general practitioner (GP) treatment of problem drinkers
The
assessed in a randomised controlled trial. 40 problem drinkers referred consecutively to a specialist alcohol clinic by their GP were, after assessment, randomly allocated to either GP or specialist clinic treatment groups. All subjects received intitial advice and counselling in the clinic about their drinking. The specialist clinic group received continued care from the clinic including, if necessary, admission to hospital. Patients in the GP group were returned to the care of the G P who was contacted and supported by the specialist. After 6 months of follow-up, there were significant reductions in alcohol consumption and alcohol-related problems in both groups. No significant difference was found between the two groups with respect to the main outcome measures. No differential treatment effect was found with the more severely dependent drinkers. The findings show that after an initial detailed assessment and advice session, the treatment provided by GPs is at least as effective as that from a specialist clinic with respect to improvements in drinking behaviour and alcohol-related problems. After initial assessment and advice, specialist clinics should encourage GPs to become more involved in the subsequent care of problem drinkers. Such a practice should be based on the individual patient’s needs and the adequacy of support offered to GPs. was
Introduction The role of the general practitioner (GP) in the identification and treatment of problem drinkers has been the focus of much interest and debate.l-6 The Kessel report’ suggested
that GPs were ideally placed to provide treatment and care for this population and gave impetus to the development of community-based services for problem drinkers in the 1980s. However, GPs seem to be reluctant to take on this extra burden.’-1 Lack of resources and support, inadequate education, and negative views about treatment efficacy have
been blamed for this disinclination. The concept of the Community Alcohol Team has been criticised," but the specialist support and training which it provides has been shown to have a beneficial effect on GPs’ therapeutic commitment to problem drinkers.12 Although advice and information given to heavy drinkers by GPs reduces alcohol consumption,13 is the treatment provided by GPs as effective as that from a specialist clinic? This important question has not previously been studied in a controlled trial. Furthermore, it is not known whether patients benefit from specialist or GP treatment differentially according to the extent of dependence or problem severity. Such a finding could have implications for appropriate treatment
matching. The study reported here was conducted as part of a larger investigation of the attitudes and behaviour of the providers and recipients of GP and specialist treatment of drinking problems (unpublished). The null hypothesis of this controlled trial was that GP and specialist clinic care were equally effective. Orford and colleagues14 found that severely dependent drinkers given intensive specialist treatment had a better outcome than did those given simple advice. A further hypothesis of our trial, therefore, was that severely dependent subjects would benefit more from specialist clinic treatment. Patients and methods Patients 40 problem drinkers, referred consecutively to the Maudsley Hospital Alcohol Clinic by their GPs, were randomly assigned to one of four specialists (one clinical nurse specialist and three psychiatrists). All patients received a full clinical assessment, including psychiatric and physical examination, and relevant blood investigations. Also, patients were asked to complete the Severity of Alcohol Dependence Questionnaire (SADQ), the General Health Questionnaire (GHQ), and the Alcohol Problems Questionnaire ADDRESSES Addiction Research Unit, National Addiction Centre, Institute of Psychiatry, and Maudsley Hospital, Denmark Hill, London, UK (D. C. Drummond, MRCPsych, B. Thom, MA, C Brown, RMN, Prof G Edwards, FRCPsych, M. J. Mullan, MB) Correspondence to Dr D C Drummond Addiction Research Unit, National Addiction Centre, Institute of Psychiatry, 101 Denmark Hill, London SE5 8AF, UK
916
(APQ). Informed consent was obtained and patients were randomly allocated to one of two treatment groups-namely, the clinic group and the GP group. Subsets of patients defined by a SADQ score (two levels) were separately randomised to maximise the likelihood of group matching. The clinic group received counselling and advice about their drinking problem followed by routine outpatient clinic care and, if necessary, admission to hospital. The GP group were also counselled and were given advice about their drinking problem, but after this initial consultation, they were returned to the care of their GP. These patients were assured that the specialist would remain involved in their treatment via contact with their GP and they were told to contact their GP for an appointment. Specialists then arranged to visit the GP to discuss the management of the patient, and they provided a short booklet of guidelines on the management of alcohol problems. The specialist offered further advice and support to the GP, initiated further contact by telephone to check on progress or difficulties, and provided a contact telephone number. Although patients in the GP group were encouraged to attend the GP rather than the clinic, a minority decided to attend the Maudsley and other specialist clinics during
follow-up. The initial assessment interview, including counselling, took a of 3 h. All patients and GPs were told that they would be contacted after 6 months to assess outcome. Follow-up interviews with patients were conducted by two research workers who were not blinded to the patient’s allocated group but who were not connected with the patient’s care. This interview sought information about alcohol consumption and related problems, treatment contact, and perceptions of the care offered. Patients were also asked to complete the APQ, GHQ, and a questionnaire designed to measure their perception of change in problem status. mean
Measures The SADQ15 consists of a 20-item self-completion questionnaire which measures the frequency and severity of symptoms of the Alcohol Dependence Syndrome’6 during a recent period of heavy drinking. The GHQ (see Goldberg17) has been widely used in studies of the prevalence of psychiatric disorder in clinical populations. The APQ is a new questionnaire which has been TABLE I-DEMOGRAPHIC DATA AND DRINKING HISTORY AT INTAKE
alcohol-related problems of the previousf: consists of 46 items about a wide range of common!;’ months;18 it affected problem areas, although only an aggregate score of the 23 common items which were applicable to all subjects are reported here (APQC score). Alcohol consumption during the 6 months before and after the initial consultation was measured by a standardised interview method. The patient was asked to recall the maximum amount of alcohol consumed on any one day during each of the 26 week Patients were also asked to rate change in problem status at follow-up as either ’improved’, ’worse’, or ’unchanged’ on a 13-item, 3-point Lickert scale. An overall self-rated change score is the unweighted sum of the 13 variables.
designed
to measure
Results
Adequacy of matching at intake (table i) The two groups were broadly similar with respect characteristics (Student’s t-test). to various intake Nevertheless, because there was a difference between the groups in mean age and time in present accommodation, these variables were used as covariates in subsequent analyses. Since SADQ score was a matching variable, this was also used as a covariate in subsequent analyses. Follow-up Outcome. Follow-up data were obtained for 19 of the clinic patients and 18 of the GP patients. 1 patient in the clinic group and 1 in the GP group refused a follow-up interview, although information about the drinking behaviour of the patient in the GP group during follow-up was obtained from the GP. 2 GP patients could not be traced and no relevant information was available from their GPs. GHQ and APQ data were obtained for 17 and 16 cases, respectively from each group at follow-up. Since it proved impossible to interview all subjects at the 6-month followup point, the mean length of time (SEM) after initial assessment when follow-up interviews were conducted was significantly longer in the clinic group than in the GP group (10-0 months [0’8] vs 7-7 [0’5]; p
