Residential services differ from many community psychosocial services in that the former almost always exclude clients who are actively using substances either controlled or prescribed. As such, residential services will almost exclusively form part of an aftercare package, whilst community psychosocial services may cater for clients at all stages of the cycle of change.

RESIDENTIAL SERVICES

The 'Effectiveness Review' (1995) identifies four modalities of residential service:

• Therapeutic communities.
• Twelve-step Minnesota Model houses.
• General houses including those with a Christian-based philosophy.
• In-patient treatment.

The first three of these share a number of key features:

• Residents must be drug free (apart from tobacco); some services provide in-house detoxification (especially the 12-step houses).
• They provide a structured programme of psychological, educational and social therapy, which aims at preparing the drug misuser to manage better a drug-free life back in society.
• Programme lengths of between 6 weeks and 9 months (most commonly 3 to 6 months).

In-patient treatment is more medically based and provides detoxification and counselling (individual and/or group work). Lengths of stay are rarely over 2 months with an average of four weeks. Services are based in hospital psychiatric wards, specialist in-patient units and the voluntary sector. The average size of such units is 12 beds.

As for in-patient detoxification, residential psychosocial rehabilitation appears to present a higher probability of successful outcome than does community based rehabilitation; this is however, very different from saying that one setting is more or less cost-effective than the other. In the current political context it seems probable that the push to create increased treatment places with limited resources may occur at the expense of residential placements.

A 1996 follow-up study, conducted as part of the National Treatment Outcome Research Study (NTORS) reported the following (Gossop M et al, 1999):

• During 1995, 1075 drug users (as distinct from alcohol users) entered 54 residential services. 405 subjects entered the 15 residential rehabilitation units and 8 in-patient units that were included in the NTORS study. Of these subjects, 75% were heroin users, but poly-drug use was the norm.
• 275 subjects were re-interviewed one year later. At this time 37% had been abstinent from opiates, stimulants and benzodiazepines over the past three months, 19% were drinking excessively compared to 33% at intake, stimulant use fell from 71% to 32% including a halving of crack use, the proportion of those injecting was roughly halved, the proportion of those committing crime was roughly halved.
• Half of the clients did not complete the full period of residential care. Compared to early leavers, those retained at least 28 days in short programmes, and 90 days in longer ones were four times less likely to use opiates 9-12 months later; they were also far less likely to use other drugs, commit crimes, or inject. In shorter term residential units, 64% of clients stayed for the critical period. In longer term units, only 40% stayed for the critical period. In inpatient units, only 20% stayed for the critical period. Early leavers still improved, most noticeably in stimulant use and sharing injecting equipment.
• Clients of the 25% of projects with the poorest retention rates had not cut their heroin use at all.

There are several implications of the above results.

• When compared to the community services in the NTORS project, residential services achieved comparable drug misuse outcomes, but were treating clients with more severe patterns of substance misuse,