Comorbidity of severe mental illness and substance misuse is now common in general psychiatry (Reference Regier, Farmer and RaeRegier et al, 1990), and perhaps almost standard in forensic psychiatry (Reference SnowdenSnowden, 2001). It is also reflected in child and adolescent psychiatry (Reference Boys, Farrell and TaylorBoys et al, 2003) and even in old age psychiatry (Reference Jolley, Kosky and HollowayJolley et al, 2004). The range of hazards associated with substance misuse in people with mental illnesses includes elevated risk of relapse of psychosis (Reference Cantwell and HarrisonCantwell & Harrison, 1996), increased frequency of hospitalisation (Reference Bartels, Teague and DrakeBartels et al, 1993), poorer compliance with treatment (Reference Jablensky, Sartorius and EmbergJablensky et al, 1992), higher levels of treatment-resistance (Reference Bowers, Mazure and NelsonBowers et al, 1990), impairment of the integrity of therapeutic regimes in hospital settings and in hostels in the community (Reference SandfordSandford, 1995), stress in the community (Reference Drake and WallachDrake & Wallach, 1989), higher rates of homelessness (Reference Scheller-Gilkey, Lewine and CandleScheller-Gilkey et al, 1999), increased suicidality (Reference Drake and WallachDrake & Wallach, 1989), and increased potential for antisocial behaviour and crime of both an acquisitive and a violent nature (Reference Stewart, Gossop and MarsdenStewart et al, 2000; Reference Sinha and EastonSinha & Easton, 1999). The misuse of substances is therefore a significant obstruction to the effective use of psychiatric treatment, and the financial cost associated with such clinical adversity must run into millions of pounds.

The problem of ongoing substance misuse by inpatients in psychiatric hospitals is well recorded nationally and internationally, both in general psychiatry (Reference Phillips and JohnsonPhillips & Johnson, 2003) and in forensic psychiatry (Reference Dolan and LawsonDolan & Kirwan, 2001; Reference van Der Laan and JanssenVan der Laan & Janssen, 1996). Psychiatric patients, as vulnerable people, may be a particular target for exploitation by drug dealers (Reference GouldGould, 2003). The main strategies used by mental health services in addressing comorbid substance misuse have included testing for illegal substances in samples of urine (Reference WithersWithers, 2001), or less frequently hair (Reference Kelly and RogersKelly & Rogers, 1996), while potentially samples of blood, saliva, nails and breath may also be used (Reference WithersWithers, 2001); search and removal of illegal substances, or indeed of legal but harmful substances such as alcohol from patients or close supervision or even exclusion of visitors who may bring them in; removal of leave status for varying periods of time; and attempts at treatment of substance misuse by psychological and medical methods, though the successful and sustained containment of substance abuse in psychiatric in-patients has had its limitations.

For patients in psychiatric hospital, whether informal or detained, the prevention of access to illegal drugs and alcohol seems to have been generally regarded as too difficult to achieve. Even units providing medium security find themselves unable to render their environment free of illegal drugs and alcohol (Reference WheatleyWheatley, 1998; Reference Dolan and LawsonDolan & Lawson, 2001). Only in the high-security hospitals, with the benefit of high levels of physical security, have the levels of unwanted alcohol and drugs been kept to a minimum (Reference Kendrick, Basson and TaylorKendrick et al, 2002). In the absence of such high security, and indeed even as an adjunct to it (Reference Kendrick, Basson and TaylorKendrick et al, 2002), other methods of discouraging access of addictive substances could be employed.

Canine olfactory sensitivity has been estimated as being superior to that of humans by a factor of at least 6 million. King Richard I in the 12th century commented on the hound’s ability to find an escaped felon, and such tales of a dog’s potential have been recorded as far back as the year 300 bc in Roman times (Reference BirdBird, 1996). More recently, dogs have been used to detect explosives, incendiary materials used to start fires in arson investigation, detection of people trapped in damaged buildings, crowd control (Reference StittStitt, 1991), elucidation of early signs of cancer in urine samples (Reference DobsonDobson, 2003) and in sniffing for substances of misuse (Reference GuerraGuerra, 1992). Of even greater olfactory efficiency than dogs in such regard are pigs (Reference BirdBird, 1996). However, their relationship with humans being less one of domesticity has eclipsed their use for such purposes.

The first city to conduct organised police canine training in law enforcement was Ghent in Belgium in 1899 (Reference StittStitt, 1991), and subsequently it has been used in the United States, Britain and across the world. However, the negative image of fierce dogs used in subduing encaptured populations or quelling legitimate public protest has also been an enduring public memory (Reference StittStitt, 1991).

Though initially dogs used by police were selected for their more ferocious appearance, smaller canines may have better olfactory ability and are also more amenable to community acceptance (Reference FriedmanFriedman, 1984). Developing a skilled narcotics detection dog depends on appropriate dog selection followed by effective training of both the dog and its handler. Some training programmes use single dog handlers, whereas others prefer multiple dog handlers in order for the employment of a dog not to be dependent solely on the availability of only one person (Reference FriedmanFriedman, 1984). Training of dogs requires them to meet certain standards for qualification, with ongoing maintenance training thereafter. The reliability of trained dogs is very high, at least 95 per cent, being able to penetrate even the most sophisticated concealments by drug traffickers. However, a small percentage of false positives and false negatives can occur. The consequences of a false positive are the unreasonable allegation against an innocent person, while a false negative would allow the entry of illegal substances into a building (Reference BirdBird, 1996). Errors of detection are, however, more often the result of the dog handler misinterpreting the dog’s responses than the accuracy of the dog itself. The dog and its handler(s) need to be seen as a unit on which the efficacy of detection depends. An important element in efficacy may be the environment in which the dog is being used. Random sniffs of large numbers of people may lead to higher levels of false positives than in circumstances where dog sniffing is combined with advance profiling or knowledge of known drug misusers or carriers. Other potential hazards to be avoided include infliction of dog bites on innocent people, or even on culpable drug carriers, avoidance of transmission of infection by the dog, ensuring appropriate periods of relaxation of trained dogs, provision of vehicles modified to carry dogs used in drug detection, and the avoidance of dogs themselves becoming addicted to narcotics through training and detection (Reference FriedmanFriedman, 1984).

The use of drug dogs in psychiatry is to date very limited. However, they have been employed in high-security hospitals for some years (Reference Kendrick, Basson and TaylorKendrick et al, 2002). To a lesser extent, they have also been brought in on occasions in some medium secure units. More recently, the use of dogs in general psychiatric units has been felt to have contributed to suppressing the supply of illegal drugs (Reference GouldGould, 2003).

The undertaking of drug dog searches on psychiatric units, being premised on the need to create and maintain a safe therapeutic environment, precludes the obtaining of consent by patients or affording to them any advance warning that such a search is to take place. It would be unreasonable for such a search to be prevented due to refusal of consent by patients with illegal drugs in their possession. Similarly, patients cannot usually be informed in advance as that would defeat the purpose by enabling patients to remove any illegal drugs prior to the search. Patients should, however, be informed generally that drug dogs may be brought in at various times without them knowing specifically exactly when.

Staff themselves may have differing views on the use of drug dogs, though no surveys in psychiatric units have been published. Extensive discussion is therefore appropriate on units where such an approach is being introduced. Staff objections may be based on various factors including a perception that drug dogs are unfamiliar in therapeutic settings and more associated with a custodial or punitive perspective. Staff will also vary in their affability towards dogs, some liking them more than others and a few being quite aversive to them. There may also be objections based on religious factors, dogs being regarded as unclean by Muslims, while for Muslims and orthodox Jews, the pig would be seen as highly contentious. Islam is, however, also prohibitive of the abuse of alcohol and illegal drugs and as such would probably favour an environment uncontaminated by such substances. Nonetheless, measures to ameliorate such concerns can include placing sheets on the floor of the area searched by the dog, while the modern search method is of a passive nature, the dog being trained to sit when it smells a suspected substance and to avoid touching people. Prior to prayer, Muslims would generally be encouraged to wash, especially if any contact with a dog had occurred. Staff in psychiatric units are generally amenable to seeing the benefits of dog searches as a more effective strategy in eliminating illegal drug supply on units with all its adverse consequences.

Operationally, the dog search should be used not only to detect if drugs are present but also as a further opportunity for staff to discuss with patients the hazards of such substances to patients in regard to effects on their mental state and behaviour, prospects for discharge and the need to refrain from substance misuse within the community. Many patients will see the benefit of removing the drug culture from the ward.

While the dog search is in progress, patients should be asked to be present when their rooms are entered, their consent should be sought and many will give it, but the search is not contingent on it. Any suspected drugs found are placed in a container and documented, although a sniff may sometimes indicate past presence of illegal drugs in that location though none are any longer present. When illegal drugs are found, patients need to be informed and asked by staff if they can account for the finding. As it is illegal to be in possession of such drugs, there may be cases where positive finds may in due course result in criminal charges being brought, in which instance the patient should be advised to obtain legal advice. Patients generally also need to be informed that they have access to the hospital complaints procedures or indeed any other appropriate agency should they feel that the drug search was conducted improperly. Where patients’ visitors are suspected of having possession of illegal drugs, the visitor may be asked to agree to a search and be denied access if he declines to consent to be searched.

As detoxification may be an important component in the treatment of drug and alcohol misuse (Reference LutyLuty, 2003), the creation of a drug-free environment in hospital is a clinical necessity. Units providing treatment for people with mental illnesses are also required to provide a safe environment (Department of Health, 1998), the degree of which is clearly compromised by the effects of active substance misuse. In addition to the provision of effective treatment of substance misuse, the prevention of entry of substances into psychiatric hospitals, and related community provision, can only be beneficial. In the future, psychiatrists and related health professionals may yet find that the dog is man’s best friend, and may find it necessary to include the dog (or perhaps even the pig) as part of the clinical team!