Comorbid substance misuse in acute psychiatric inpatients

You have to be extremely unwell to be admitted to an acute psychiatric ward as the number of beds shrink year on year. Using substances is one factor that is associated with requiring hospital admission rather than community treatment. As Hermine Graham, the lead author of a recent study points out, admission can be a ‘window of opportunity’ to explore and discuss substance use with the client as a trigger factor that led to hospital admission (Graham et al, 2015).

Effective interventions for people with co-existing acute mental health and substance use problems have proved elusive so far. There have only been a few trials to date, as research funding has been difficult to secure. In part due to the high attrition rate of this client group in trials. This loss to follow up will be a familiar factor in clinical practice where it can be challenging to engage and retain these clients in treatment. So the authors are to be admired for investigating what might work with these clients.

As fellow Elf Sally Adams noted last year in a blog about alcohol and depression, there has been a disproportionate focus in research on mild to moderate levels of depression and alcohol use (Adams, 2015).

Despite significant need, we don’t know what is effective for people with comorbid mental illness and substance misuse.

Methods

The Graham et al (2015) study was a small scale feasibility trial to establish whether the intervention was practical and acceptable.

New admissions to eleven acute wards and three psychiatric intensive care units in one English mental health trust were recruited into the trial
The participants were randomly allocated to treatment as usual consisting of inpatient assessment and monitoring by staff on the ward. The intervention group received Brief Integrated Motivational Intervention (BIMI), consisting of a blend of cognitive behavioural and motivational approaches
The BIMI intervention was delivered over two weeks in a series of 4-6, 15-30 minute sessions. A booster session was delivered 1 month later
A power calculation was used to inform the sample size, the investigators aimed for a 1 point difference on the Substance Abuse Treatment scale (SATS) as the primary outcome measure
The researchers were blind to participant treatment group until the 3 month follow up
59 patients were recruited from 14 wards
Alcohol and cannabis were the two substances most commonly used.

Results

Overall there was little difference between the groups. The primary outcome of engagement showed only a modest improvement, statistically discerned rather than being clinically significant.

Increased contact with staff was found in the BIMI group, but this did not produce any significant change in the outcomes measured during the short duration of the trial, when compared to the control group.

Both groups reported using their primary substance less often in the last month of the trial period.

All trial participants used substances less frequently.

Conclusions

This is an important area to research, but as this study demonstrates it is not easy. Illustrated by the 30% (n=9) of the intervention participants unable to receive BIMI as they were either discharged from hospital or refused.

As the authors note, baseline assessment can produce an effect on its own. In other words, just asking someone about their substance use can instigate a positive change. Such a simple thing to do, surely inpatient staff have the time to ask this one question?

Simply asking about substance use produces positive results.

Strengths and limitations

Strengths

The investigators focussed on an intervention for acute inpatient care
The intervention is low cost

Limitations

The authors rely on 10 year old data as an indication of how prevalent substance use is for inpatients
The intervention staff received 2 days of training, worked with specially trained dual diagnosis staff and received supervision from the investigators
The exclusion criteria is not clear, so we don’t have much detail on who is potentially missing from the trial or why
It is likely that clinicians would be able to guess which arm of the trial participants were in during the trial introducing the possibility of bias
Very few females were included in the trial (n=9)
Tobacco use was not studied.

Summary

The results may seem disappointing, but this is a useful reminder of how challenging research in the speciality of co-occurring substance use and mental health is. The findings of this trial are similar to those of Barrowclough et al (2010) who used comparable interventions with this client group and only found a modest reduction in alcohol use despite an intensive treatment approach.

A national survey is urgently needed to establish the extent of co-existing substance use in this population. This should gather information on the type and frequency of substance use, this information would help inform services in training their staff and providing appropriate interventions.

It is vital that substance use is not only assessed for this client group, but that staff have the confidence and skills to act on this information once they have it.

We need to know more about how and why patients use substances.

Links

Primary paper

Graham HL, Copello A, Griffith E, Freemantle N, McCrone P, Clarke L, Walsh K, Stefanidou CA, Rana A, Birchwood M. (2015) Pilot randomised trial of a brief intervention for comorbid substance misuse in psychiatric in-patient settings. Acta Psychiatrica Scandinavica Article DOI: 10.1111/acps.12530 [Abstract]