Reliability of earlier editions of the PSE
Earlier version of the PSE have been extensively tested in a wide range of working conditions. An early scepticism, reasonably based on studies showing that free clinical interviews did not produce comparable diagnostic results from different examiners interviewing the same series of patients (Kreitman, 1961; Zubin, 1967), was dispelled by data using PSE-3-PSE-5 (Wing et al., 1967), which demonstrated high concordance between interviewer–observer pairs and between different interviewers examining the same patient. Kendell and colleagues (1968) using PSE-7 confirmed these results. The same scepticism had to be overcome when choosing an instrument for the USA–UK Diagnostic Project and the International Pilot Study of Schizophrenia. PSE-7 and PSE-8 were successfully and reliably used in both projects (Cooper et al., 1972; WHO, 1973; Wing et al., 1974). The experience did demonstrate, however, the need for more precise definitions and instructions when examiners from a wide range of backgrounds were collaborating in an international study. Luria and colleagues (1974, 1979) also demonstrated that the system could be used reliably in the United States.
PSE-9 has been further tested in numerous projects throughout the world. Reliability of scores and classifications has been found to be good in a wide range of conditions, including general population samples and most types of clinical setting (Bebbington et al., 1981; Cooper et al., 1977; McGuffin et al., 1986; Pakaslahti, 1987; Rogers and Mann, 1986; Sturt et al., 1981; Wing et al., 1977; Wing et al., 1978). A full discussion of the concept of reliability as it affected the PSE-9 used in a follow-up of the IPSS series has been provided by Gulbinat (1979).
The prototype for SCAN
The Chart in Chapter 2 shows the progression of SCAN through four consecutive versions: 0, 1, 2.0 and 2.1. The first working draft of a prototype (not then named) was dated September 1983. It incorporated most of the changes in design described in Chapter 2, and was adopted following pilot interviews with short-term and long-term patients in a variety of hospital and community settings.