Identification and management of alcohol use and illicit substance use in outpatient psychiatric clinics in Sweden: a national survey of clinic directors and staff

It is well known from epidemiological studies that people with a psychiatric disorder frequently have a concurrent substance abuse or dependence concerning either alcohol or illicit substances [1‐3]. Prevalence of substance abuse or dependence among persons in the general population with anxiety or depression is estimated to be 25–30%, with markedly higher levels among those with more severe psychiatric disorders such as schizophrenia [1]. The prevalence is believed to be just as high among patients in psychiatric clinics [4‐7]. Individuals with concurrent psychiatric disorder and substance use problems have a worse treatment prognosis [8] and an increased risk of later relapse in their psychiatric disorder [9]. Hazardous alcohol use, a drinking pattern not deemed to be a fully developed alcohol abuse or dependence but with potential to lead to adverse consequences, is also problematic in a psychiatric setting. Excessive drinking commonly interferes with psychosocial functioning and raises the risk of subsequent escalation of alcohol problems. In fact, even moderate alcohol intake has a negative impact on clinical course and response to treatment [10, 11], and may interact negatively with common psychiatric medications such as fluoxetine [11] and benzodiazepines [12]. Reduced hazardous drinking among psychiatric patients has been associated with more rapid symptom improvement in anxiety and depression [13]. Hazardous illicit substance use is not an equally established concept, but it is not controversial to suggest that sporadic use of illicit substances also has negative implications for treatment and recovery from psychiatric disorders.

The importance of detecting both hazardous alcohol use and abuse or dependence among patients in health care settings outside specialized addiction care was established by the World Health Organization with the development of the Alcohol Use Disorders Identification Test (AUDIT), a questionnaire developed specifically to facilitate identification of alcohol problems within primary care [14]. This opportunistic approach to identifying hazardous alcohol use as well as abuse and dependence has subsequently been incorporated into a host of national guidelines such as those issued by the National Institute for Health and Excellence (NICE) in the UK [15], the National Institute on Alcohol Abuse and Alcoholism (NIAAA) in the United States [16] and the National Board of Health and Welfare in Sweden [17]. These guidelines propose that health care staff should routinely carry out alcohol screening, preferably with a validated questionnaire such as the AUDIT. When alcohol abuse or dependence is identified, the patient is to be offered referral to appropriate treatment, and when hazardous use of alcohol is identified, this should be addressed immediately within the current health service with structured feedback, preferably by means of a brief intervention (BI). BI is a single-session treatment aimed at helping hazardous drinkers moderate their drinking by providing screening, feedback and brief advice regarding their alcohol consumption. BI has been shown to have small but stable effects in the primary care setting [18] and in the emergency department setting [19]. Although research on BI in the psychiatric setting is scarce, a few randomized trials have demonstrated that, although efficacy in terms of effects on alcohol consumption is not yet clear, BI is at least feasible and safe to use in this population [20‐23]. In Sweden, some progress has been made in evaluating the implementation of national guidelines for substance abuse and dependence care through quality indicators from the National Patient Registry and the National Quality Registry for Dependency (SBR), but so far no systematic information on screening and BI within psychiatry is available [24]. The importance of detecting illicit substance use has not been placed on the wider healthcare agenda to the same degree as alcohol, but the Drug Use Disorders Identification Test (DUDIT) [25, 26], a parallel instrument to the AUDIT, is nationally recommended by the National Board of Health and Welfare in Sweden for identifying illicit substance use in a variety of health care settings as well as within specialized addiction care [27]. In the UK, the 2011 NICE guidelines on coexisting severe mental illness (psychosis) and substance misuse recommend assessment and management of illicit substance use in wider healthcare settings, specifically stipulating that individuals seeking help within psychiatry should not be excluded due to illicit substance use [28]. In the US, the current focus lies more on ensuring a continuum of care for individuals with illicit substance use [29], with limited research published on identifying and managing illicit substance use within psychiatry [13]. Regarding the efficacy of BI for illicit substance use, the evidence is unclear. Few studies have been conducted on BI for illicit substance use, and a minority of these have been conducted in clinical settings. Two large randomized trials have been conducted in a primary care population: one of these identified a reduced drug use effect following BI [30], while another did not [31]. Recent pooled evidence suggests that interventions delivered via the internet could yield small, but significant overall effects beyond control conditions including BI [32]. To our knowledge, no studies have investigated BI for illicit substance use in the context of psychiatry.

Given the high levels of comorbidity and negative implications of both hazardous use and substance abuse/dependence for treatment outcomes, it would seem essential for any psychiatric clinic to have an effective strategy in place to detect and assess these problems among patients. This could be achieved by routinely screening all patients for alcohol and illicit substance use at initial assessment. Such a strategy could identify many patients early on, facilitating collaboration or referral to appropriate substance use treatment. Even better, it could also identify patients with these problems early enough to allow clinicians to offer on-site treatment such as BI or other suitable measures to address and reduce substance use. However, we have found no existing research on the extent to which screening for alcohol and illicit substance use is conducted in psychiatric outpatient clinics, nor to what extent patients with hazardous use of alcohol or illicit substances are offered BI.

The purpose of this study was to investigate current national practices for identifying and managing hazardous substance use and diagnosed substance abuse/dependence (both alcohol and illicit substances) among patients in psychiatric outpatient clinics in Sweden. Specific research questions were:

The main purpose of this study was to assess the existence of guidelines for screening and BI for substance use (alcohol and illicit substances) at psychiatric outpatient clinics in Sweden, as well as actual use of screening and BI among staff at these clinics. Almost all clinic directors who responded to the survey indicated that they had clear guidelines for staff to screen for alcohol use and illicit substance use during initial assessment, suggesting high level of awareness in psychiatric clinics about the importance of identifying substance abuse and dependence. However, only about two-thirds of staff members stated that they routinely screened for alcohol during initial assessment, and even fewer stated that they routinely screened for illicit substances. Furthermore, about half of responding clinic directors indicated that they had guidelines for offering BI, while about one-third of staff members indicated that they actually used BI regularly, suggesting that BI is not an integrated part of psychiatric treatment in Sweden.

There are several limitations to this study, most of them related to selection bias. First, with an attrition rate of about 40% among the clinic directors approached, those responding to the survey may have differed in some unknown way from those who did not respond. For example, those clinic directors who chose to respond to the survey may have had more experience of, or knowledge about, substance use and its relevance in a psychiatric context. Those who chose not to respond may have been those who knew that their clinic lacked guidelines. Secondly, only about half of clinic directors responding were willing to distribute the survey to their staff, a factor that may have introduced selection bias in the subsequent staff survey. Directors at psychosis clinics were more willing to distribute the survey, but no other differences were identified when comparing directors who agreed to distribute the survey to their staff, with those who did not. Accordingly, these two groups could have differed in some other way that was not measured. For example, those who chose to distribute the survey may have had more knowledge or experience of working with substance use issues. Also, the clinic directors who chose to distribute the survey could hypothetically have selectively distributed the survey only to a certain subsection of their staff, for example to those at the clinic whom they knew worked with or were more familiar with substance use. Thirdly, almost 60% of staff members who were estimated to have been sent the survey did not respond. We do not know to what extent staff members who responded to the survey were representative of the staff population at large. It could be that staff members who considered substance use an important issue were the ones who took the time to answer the survey, they may have had more experience in working with these issues, or they may have had more substance abuse training. A final limitation relates to the anonymous nature of the staff survey; since we did not ask staff members to indicate where they worked, we have no information about which specific clinics they were working at. There may have been some clinics where no staff responded and other clinics where all or most staff members responded.

Constructive management of patients with substance use in psychiatric outpatient clinics is a complicated matter, involving decision-making at the clinic level as well as policy decisions in the form of national guidelines over which the individual clinic has no influence. Nevertheless, awareness of a patient’s substance use habits is unquestionably a crucial component of psychiatric treatment, and psychiatric clinics should have a well thought-out plan in place on how to identify and manage patients’ substance use. The clinical research literature consistently suggests that treatment of patients with both a psychiatric disorder and substance abuse/dependence optimally should focus on both disorders, rather than on focusing on just one or the other [37]. Although psychotherapy programs have, with some success, been developed specifically to satisfy this aim [38, 39], integrated treatment is often difficult to offer under real-life conditions given the fact that psychiatry and addiction clinics usually are managed by separate health care entities. Nonetheless, about half of the clinic directors in our survey indicated that they maintained local collaboration with addiction care services.

Use of BI was quite rare in our sample, a finding which perhaps is not surprising. Although BI has been shown to be a safe and effective intervention for alcohol when used in other parts of health care, particularly in primary care [40], its place in psychiatry is still unclear and more trials are needed before any conclusions on efficacy in this population can be drawn. Further, the use of BI for illicit substance use is even less investigated. Thus, research is still lacking as to whether BI is even feasible or safe in a psychiatric population. Further, our finding that increased substance abuse training was the most common recommendation for clinical improvement by both clinic directors and staff suggests that the lack of substance abuse training among staff was perceived as a real problem by both groups. Offering staff substance abuse training could thus potentially be an effective way of increasing adherence to guidelines [41], as relatively brief training sessions have been found to improve staff members’ self-perceived knowledge and attitudes towards working with this issue [34, 42]. Lastly, given the consistently high rates of comorbidity reported among patients with schizophrenia, with up to 50% of patients at some point in their life suffering from substance abuse or dependence [1], and considering the fact that premature mortality among these patients due to medical conditions attributable to substance use is common [43], the apparent low use of screening among staff in psychosis clinics was surprising and seems to suggest a very real problem that needs to be addressed.

Almost all clinic directors stated that they had clear guidelines for staff to screen for alcohol use and illicit substance use, but use of screening among staff was markedly lower than would be expected. Our results thus suggest a gap between the guidelines that the clinic directors report and actual adherence to guidelines by staff. Substance abuse training among staff is rare and may be a contributing factor to low adherence since those indicating that they had substance abuse training reported performing screening and brief intervention to a greater degree. Provision of substance abuse training to staff in psychiatry might thus be a key factor needed to improve frequency of screening and brief intervention. Clinics specializing in psychosis might need to consider making extra efforts to improve screening practices.