Page 149 - PhD GT

base and was chosen on the basis of convenience due to resource limitations. While a convenience sample is a type of non-probability sample (Henry 1990) and such samples are characterised by a degree of subjective judgment in the selection process, such judgment was limited to selection of the period of recruitment and to the application of minimal exclusion criteria in the same way that would occur in a probability sample.
The goal was to select a sample which was representative of all help-seekers at specialist National Health Service agencies for problems of heroin and alcohol use and dependence. The sample was somewhat biased towards those patients who attend the central agency site for at least their first appointment as opposed to those who were seen in their own homes or other community venue. The possible basis for such bias is discussed in the section below.
7.1.1 The research site
The agency chosen for the study is a National Health Service out-patient and day-patient facility with six hospital beds (since closed) at a separate site. Detoxification programmes and psycho-social treatments of substance dependence are provided by three clinical teams under the overall management of a consultant addiction psychiatrist. The agency is a specialist service provided by the Leeds Community and Mental Health NHS Trust. Each clinical team contains a multi-disciplinary staff group with medical staff (mainly psychiatrists and a general practitioner), psychiatric nursing staff, a psychologist, social worker and addiction therapists. In addition to the clinical teams, a training team provides a series of University courses and short courses for those professional groups engaged in addiction work in the specialist and generalist sectors; an Information and Research team operate the Regional Drug Misuse Database (Department of Health 1989, 1996) and maintain the Unit databases for purposes of clinical evaluation and audit.
143