OUP user menu


Preben Bendtsen, Ingemar Åkerlind
DOI: http://dx.doi.org/10.1093/alcalc/34.5.795 795-800 First published online: 1 September 1999


During an intervention period of 1 to 2 months, a project team supported general practitioners (GPs) and nurses in four primary health care centres in Sweden in introducing new routines for detection and treatment of problem drinkers. After the implementation of the new methods, the GPs reported increased involvement in early detection and intervention significantly more often than the nurses did. A majority in both groups reported perceived improvement in skills. There was a significant positive change of the attitudes concerning working with alcohol-related problems in the nurses reaching the same level as the GPs. In the nurses, attitudes and self-perceived intervention skills were improved, but to a lesser extent than their practice. The results indicate that future efforts concerning improvement of primary health care staff involvement in alcohol interventions should focus on training, supervision, and giving positive examples, rather than on changing an already positive attitude towards alcohol intervention. The potential role of nurses is still uncertain and not utilized sufficiently.


Alcohol-related health problems are over-represented among patients seeking health care, compared with the population at large. This reflects the adverse consequences of excessive alcohol consumption on physical and mental health. The prevalence of patients with alcohol-related health problems varies in the literature. Thus, the rate appears to be up to 20% among primary care patients and between 20–40% for hospital in-patients (Persson and Magnusson, 1987; Olfson and Braham, 1992; Searight, 1992; Nielsen et al., 1994; Allen et al., 1995; Cherpitel, 1995a,b; Rambaldi et al., 1995, 1996; Wallerstedt et al., 1995).

The great majority of the population, both with and without alcohol-related health problems, seek treatment for all kinds of medical conditions in primary health care (PHC) within a 12-month period (Wallace et al., 1988; Searight, 1992). Therefore, the PHC setting is a natural focus for screening for alcohol-related health problems (Magruder-Habib et al., 1991; Searight, 1992). However, despite the high rate of alcohol-related health problems in the PHC, the majority remain undetected (Olfson and Braham, 1992; Searight, 1992; Townes and Harkley, 1994; Adams et al., 1997).

Patients respond positively to life-style-related advice, and appropriate tools for screening and intervention for patients with alcohol-related problems are available for general practitioners (GPs) (Townes and Harkley, 1994; Rickmond et al., 1996). Brief intervention of around 5 min can significantly decrease alcohol consumption (Wallace et al., 1988). Still, the current effectiveness as evaluated by the physicians themselves is substantially less than the potential. Reasons for this gap may be related to a lack of education and clinical supervision in how to detect and manage alcohol-related problems, including lack of specific counselling material (Warburg et al., 1987; Roche, 1990; Searight, 1992; Rush et al., 1994; Townes and Harkley, 1994).

Medical schools have been estimated to devote less than 1% of the total teaching hours to substance abuse. This explains why so many physicians feel inadequate when treating alcohol-related problems (Searight, 1992; Rush et al., 1994). In a previous study, less than half of 312 GPs thought that they were adequately equipped to treat alcohol-related problems (Anderson, 1985). On the other hand, a positive relationship was found between levels of training and active involvement in alcohol intervention in a recent study of 218 GPs (Adams et al., 1997).

There ought to be a continuum of interventions that match the natural development process of alcohol problems, from the primary care level to the specialist in alcohol dependence and abuse. Brief intervention methods have been suggested as useful in a PHC setting aimed at problem drinkers in an early phase (Anderson, 1993b). Barriers to implementation of such methods include attitude factors as well as lack of training of staff and information material for the patients. In addition, another limiting factor could be a lack of positive reinforcement due to the relatively low individual success rates normally experienced when treating patients with a more long-standing and advanced history of alcohol dependence (Anderson, 1993b).

The present study was therefore undertaken in order to evaluate changes in attitudes and practices among both GPs and nurses, after implementation of a training and supervision project concerning early identification and intervention of alcohol-related problems.


The study was undertaken in four PHC centres in the county of Östergötland in southern Sweden. A total of 19 GPs and 30 nurses, constituting the great majority of all staff members, answered a questionnaire before the implementation of a programme concerning early identification and intervention in a PHC setting. Four months after implementation, the participants were asked to answer the same questionnaire, complemented with some additional questions. In the follow-up 12 GPs and 25 nurses agreed to participate.

All of the nurses and 35% of the GPs were women. The median age and range was 51 and 19 years for the GPs and 48 and 31 years for the nurses. Both groups had served around 14 years in PHC centres. The non-responders were lost to follow-up, mainly because of lack of time or due to leave or change of employment at the time of follow-up. The group lost to follow-up did not differ significantly from the followed-up group with regard to sex, age, and years of service.

The intervention programme was aimed at introducing early identification and brief intervention for problem drinkers. A project team of three persons supported the PHC centre for 1–2 months, during which time both GPs and nurses were taught how to detect and advise problem drinkers. Furthermore, the project staff supervised the screening and intervention performed by the health centre's own staff. Specially designed information material for both staff and patients was produced by the project team. The assessment tools and advice about alcohol consumption used in our programme were similar to those included in several other studies (Wallace et al., 1988; Romelsjö et al., 1989; Magruder-Habib et al., 1991; Anderson and Scott, 1992; Fleming et al., 1992; Anderson, 1993a). The screening methods used were the CAGE and Mm-MAST questionnaires as well as the period-specific normal week of alcohol consumption (Searight, 1992; Romelsjö et al., 1995; Seppä et al., 1995; Wenrich et al., 1995).

In order to evaluate the change in attitudes and practices after the introduction of the early detection and brief intervention programme, a questionnaire was adapted from Romelsjö (1986).

The questionnaire applied before the intervention consisted of 38 questions divided into five categories. The first 10 questions covered background information concerning the respondent and his/her patient population. Next, they were asked to rate their current evaluation and intervention routines in four questions. Then they rated their own knowledge and perceived capacity to detect alcohol-related problems in four questions. Next, 16 questions explored attitudes and beliefs about the role of primary care in identifying and treating alcohol-related problems. Finally, they were asked to describe the current methods in use at the health centre with regard to intervention aimed at alcohol-related problems (three questions) as well as to rate their overall attitude to the possibilities of influencing the patients' drinking habits at the health care centre (one question). All ratings as well as the answers on the attitude questions were recorded on five-point Likert-type scales.

After the intervention, a modified questionnaire was used containing the same items, apart from some background questions, and with six additional questions covering specific changes in practice and skills after the introduction of the project. These specific questions were rated on four-point Likert-type scales. The questionnaires used in the present study can be obtained from one of us (P.B.).

SPSS software version 8.0 was used for statistical analyses. In order to test significant changes in practices between the GPs and the nurses, the Mann–Whitney U-test was applied. For analyses of changes in specific attitude items before and after the implementation of the project, the Wilcoxon signed ranks test was applied. General attitude changes were studied by t-test analyses of mean attitude score differences (least favourable response-alternative = 1 point; most favourable response-alternative = 5).


The programme had a more marked impact on GPs than on nurses. Thus, although one-fifth of the nurses reported some increased involvement in screening and counselling after the project was concluded, the great majority had not changed their practices (Tables 1 and 2) By contrast, the GPs reported a significantly more increased involvement in early detection and intervention than the nurses.

View this table:
Table 1.

Change in frequency of evaluation of alcohol consumption after the implementation of the intervention programme

FrequencyGPs (n = 12) (%)Nurses (n = 25) (%)
The difference in change between the general practitioners (GPs) and the nurses was significant (P = 0.01, Mann– Whitney test).
Much more often8.320.8
More often75.04.2
No change16.770.8
More seldom04.2
Much more seldom00
View this table:
Table 2.

Change in frequency of information to patients regarding possible relationship between symptoms and alcohol consumption after the implementation of the intervention programme

FrequencyGPs (n = 12) (%)Nurses (n = 25) (%)
The difference in change between the general practitioners (GPs) and the nurses was significant (P = 0.013, Mann– Whitney test).
Much more often8.30
More often66.726.1
No change25.073.9
More seldom00
Much more seldom00

In a specific question at the time of follow-up, most participants (60% of GPs and 50% of nurses) reported ‘some’ improvement in their skills concerning identification and intervention in problem drinkers. Furthermore, around 25% of both GPs and nurses reported a greater improvement (‘to a great extent/very great extent’). Furthermore, GPs reported more improvement than did nurses in communication skills concerning screening and intervention in problem drinkers. However, the differences were not significant. Thus, only 10% of the GPs, but nearly one-half of the nurses, indicated that their communication skills concerning alcohol had not improved.

In the prospective comparison of ratings before and after the intervention, only one of the four questions exploring knowledge and perceived skills revealed significant change in the nurse group. Among the physicians, no significant changes were seen. Thus, 32% of the nurses indicated that it was easy/rather easy to bring up the patient's alcohol habits in an examination setting after the project was implemented, in comparison to 20% before the start (P = 0.05).

Only two of the 16 questions concerning attitudes to working with early identification and intervention revealed significant changes after the implementation of the programme. More GPs (75% vs 20%) and nurses (50% vs 25%) agreed that ‘it is possible to change people's alcohol habits' (P = 0.013). In addition, the nurses (36% vs 18%) agreed that most ‘patients do not react negatively’ when asked about alcohol habits (P = 0.01). After the implementation of the programme there was a significant improvement of the composite mean attitude score in the nurses from 3.2 to 3.5 (P = 0.001), but not in the GPs (unchanged at 3.5). Thus, in contrast to the initial assessment, there was no significant difference between GPs and nurses, after the intervention, concerning attitudes towards early detection of and intervention in alcohol-related problems.


Because screening and treating alcohol-related problems are not integrated into PHC, attitudes, skills, and practices of GPs and nurses are often questioned with regard to their appropriateness (Olfson and Braham, 1992; Rush et al., 1994). The present study therefore provides an up-to-date report on the prevailing attitude, skills, and practices among both GPs and nurses, in four different health centres in Sweden.

In a previous study, physicians rated their current effectiveness with alcohol problems as substantially less than satisfactory. One of the main obstacles was the lack of training (Adams et al., 1997). A lack of training and experience in screening and intervention was also observed in our study, since the project clearly changed the GPs' and nurses' perception of their own skills by the time of the follow-up. The study underscores the need for continuous training of both GPs and nurses into screening and intervention in today's health care service.

It has previously been shown that the effects of assessment and advice are similar irrespective of whether they are applied by a nurse or a physician (McIntosh et al., 1997). In our intervention programme, both nurses and GPs participated in the education and clinical supervision and were expected to implement the skills learned in their daily contact with patients at the health centre. Before the implementation of the project, we found a more favourable attitude among GPs, than among nurses, with regard to working with alcohol-related problems. At the time of follow-up, the GPs had not changed their composite mean attitude, whereas nurses significantly increased their mean score to the same level as that of GPs. However, one particular item in the attitude scale revealed a significant change in a positive direction in both GPs and nurses. This item ‘it is possible to change people's alcohol habits', might represent an important stumbling block that needs to be overcome before the implementation of a screening and intervention programme for alcohol problems. Underlining this, several previous studies have emphasized the importance of positive attitudes from staff members towards the feasibility of asking about alcohol habits (Searight, 1992; Rush et al., 1994; Adams et al., 1997).

However, improved skills and a positive attitude are not the only factors needed in order to implement a screening programme for alcohol problems. Despite a positive change in skills and attitudes among nurses, the nurses' practice, unlike that of GPs', did not particularly increase. Thus, as many as 70% of the nurses did not change their daily routine concerning how often they evaluated alcohol consumption in their patients or gave advice. By contrast, a majority of the GPs indicated that they more often included evaluation about alcohol consumption after the project. Moreover, it must be emphasized that, before the implementation of the project, the nurses were much less involved in alcohol intervention, in comparison with GPs. Partly in contrast to our results, a previous study has shown in GPs, but not nurses, a significant relationship between having been supervised in a clinical setting and active involvement in alcohol intervention (Warburg et al., 1987).

Various reasons can be suggested for this difference in involvement between GPs and nurses. The nurses' roles and contacts with the patients are, in a Swedish health-care setting, often quite different from the GPs', perhaps involving barriers to conversations about alcohol. Whereas GPs have the main responsibility for patient examination and diagnosis, the nurses have more practical responsibilities based upon the physicians' conclusions and prescriptions. Furthermore, it could be difficult for a nurse to question a GP's diagnosis and prescription if the nurse suspects excessive alcohol consumption to be an underlying factor for the actual health problem. Another reason for this difference in the educational effect of the programme could be the fact that all nurses were women and the majority of GPs were men. Since most patients with alcohol-related problems are men, the nurses might experience difficulties in bringing up a personal and sensitive matter, such as alcohol use. This needs to be further explored in future studies.

Lastly, the lack of change in practice reported by the nurses could also indicate a different educational need and approach for nurses, than for GPs. This question was examined in various items in the questionnaire applied in the study. Thus, both GPs and nurses assessed the impact of the programme on their perceived skills, with respect to detection and intervention, as fairly similar. However, half of the nurses clearly indicated that the programme did not change their ability to communicate about alcohol habits. There seems to be a clear difference between theoretical skills or knowledge and the ability to communicate about alcohol-related matters with the patients. Again, this could reflect the special role of the nurses at the health centre, as stated above. Thus, there seems to be a need for change in the professional role of nurses, if their potential for early intervention in problem drinking is to be fully realized. In Sweden, nurses constitute a significant proportion of total staffing at PHC centres. In their daily routines, nurses have many encounters with patients where there are opportunities to talk about alcohol-related health problems. We are planning a new course aimed specifically at nurses, taking into consideration their special situation and role at the PHC centre, looking for ways of utilizing them as a resource in screening and intervention for alcohol problems. However, more research is needed in this area.

The main limitation of the present study is the relatively small numbers of participants, especially in the GP group. A larger number of participants might have revealed additional significant changes in some areas as a consequence of the intervention. Furthermore, we did not examine whether the reported changes in alcohol intervention were actually performed by the staff. A relative strength, however, is the prospective design, although the follow-up period was only 4 months. One important issue for future research is the sustainability of projects like the present one. Is there a need for a continuous training and supervision programme in PHC for early intervention in alcohol-related problems?

In general, both GPs and nurses in PHC centres have a positive attitude towards working for early detection and intervention in patients with excessive alcohol consumption. This is an important finding, since the PHC setting provides multiple opportunities for health care professionals to detect and treat alcohol-related disorders (Drummond et al., 1990; Magruder-Habib et al., 1991; Allen et al., 1995). The results indicate that future efforts concerning improvement of PHC staff involvement in alcohol interventions should focus on specific training, supervision and giving positive examples, rather than on changing an already positive attitude towards alcohol intervention. The potential role of nurses is still uncertain and not sufficiently utilized.


We wish to thank nurse Anders Larsson for his valuable help in collecting the material.


  • * Author to whom correspondence should be addressed.


View Abstract