Computerised delivery of cognitive-behavioural therapy for psychiatric disorders is developing rapidly. Self-treatment computer systems vary greatly in the degree to which they take on the therapeutic role, decreasing the need for clinician input (Reference Oakley- Browne, Toole and AndrewsOakley-Browne & Toole, 1994). At one end of the spectrum are basic aids to therapy, to be used by the clinician and patient to aid exposure in phobic anxiety, for example, computer videoclips of spiders, virtual reality depictions of heights, etc. (Reference HassanHassan, 1992). A few systems are closer to becoming complete self-help systems carrying out most of the therapeutic tasks involved in treatment, decreasing the need for clinician input by 80-95%. Such computer systems help service users to detail their problems, draw up a day-to-day treatment plan specific to their needs, rate their progress, practise coping with setbacks, do relapse prevention and recruit relatives as co-therapists if needed. Such systems (e.g. BTSTEPS, COPE, FEAR-FIGHTER, Beating the Blues, Feelbetter) need only initial screening of the service user by a clinician and an introduction on how to use the system. A number of computer treatment systems are in clinical service at the Maudsley Hospital, London, for example, BTSTEPS (Reference Marks, Baer and GreistMarks et al, 1998) for the treatment of obsessive-compulsive disorder and FEARFIGHTER (Reference Shaw, Marks and TooleShaw et al, 1999) for the treatment of phobic anxiety disorders. Others, such as Beating the Blues (Reference ProudfootProudfoot, 1999), COPE (Reference Osgood-Hynes, Greist and MarksOsgood-Hynes et al, 1998) and Feelbetter (Reference ClarkeClarke, 1999) for the treatment of depression, are undergoing randomised controlled trials. At present there are three main modes of delivery of computer-aided care: internet access (Feelbetter), telephone-accessed interactive voice response systems (COPE and BTSTEPS) and free-standing computers (FEARFIGHTER and Beating the Blues). Users feel that computer responses are empathic and understanding (Reference WeizenbaumWeizenbaum, 1976; Reference Ghosh, Marks and CarrGhosh et al, 1988) and that non-human instructions can be as therapeutic as human ones (Reference Ghosh, Marks and CarrGhosh et al, 1988; Reference Selmi, Klein and GreistSelmi et al, 1990; Reference Schneider, Schartz and FastSchneider et al, 1995). To date, however, no survey has been undertaken to look at potential service users' preferred method of delivery of computer-aided therapy.

The study

A small article was placed on the BBC teletext community pages to advertise the Maudsley Hospital's computer-aided self-help services for obsessive-compulsive disorder and agoraphobia. Information on the service and a survey questionnaire were sent out to 326 enquirers who disclosed their address. The questionnaire assessed how potential service users would want to access self-help psychotherapy for obsessive-compulsive disorder and phobic anxiety disorders. A simple tick box questionnaire was devised asking service users:

1. (a) whether they would access self-help therapy if they had to have a general practitioner (GP) referral;

2. (b) whether they wanted access to therapy via the internet, a telephone interactive voice recognition (IVR) system, CD-ROM on their home computer, a computer in their GP's surgery, a computer in a community mental health resource centre, a computer in a leisure centre/café/pharmacy, a book or another mode of access;

3. (c) how much they would be willing to pay for each computer therapy session.

Findings

One hundred and thirteen (35%) completed questionnaires were returned. Twenty-seven per cent of respondents did not want to go via their GPs to receive treatment for their anxiety disorder, and the majority of these respondents (63%) gave the reason that this process was ‘bothersome’. Thirty-four per cent wanted to access therapy via the internet, 56% via a telephone IVR system, 43% via CD-ROM on their home computer, 23% via a computer at their GP's surgery, 22% via a computer at their local community mental health resource centre, 16% via a computer in a leisure centre, café or pharmacy and 62% via a book (see Table 1). Other methods of service delivery requested were: telephone support from a human therapist (2%); audiotape or CD (3%); video (2%); group therapy (1%); face-to-face therapy (3%) and interactive television (1%). Only 9% of respondents did not want to access self-help therapy via a computer system. For therapy accessed via a computer, participants were willing to pay a mean of £10 per computer session (range 0-100).

Table 1.

Preferred method of delivery of self-help psychotherapy

	Frequency	%
Internet	39	35
Telephone interactive voice recognition system	63	56
CD-ROM on home computer	48	43
Computer at general practitioner's surgery	26	23
Computer in mental health resource centre	25	22
Computer in leisure centre/café/pharmacy	18	16
Book	70	62

Discussion

The response rate of the study (35%) while low, is average for postal surveys (Reference PartenParten, 1950; Reference KerlingerKerlinger, 1973), but does mean conclusions need to be drawn cautiously. The fact that in the initial teletext article computerised self-help services were outlined means that the study's findings may be skewed towards computer-literate respondents. However, while the majority of respondents (62%) wanted access to self-help therapy by the traditional method of a book, these same respondents wanted access via some form of computer system (91%). The authors believe this to be the first survey to ask potential users about their preferences for the delivery of self-help psychotherapies. The survey's findings have important implications for the future delivery of psychotherapy services. At present the demand for psychological therapies exceeds availability and computerised therapy has the potential to meet this need, saving on therapist time by 80-95%. The survey's findings indicate that a significant proportion of potential users do not wish to go via their GP to access such therapy (27%) and the great majority want to access self-help therapy via a computer system (91%), more specifically a system which can be accessed from home (82%) (the internet, telephone IVR system and CD-ROM on their home computer). These findings must be kept in mind in the future development of computerised psychotherapies.