The surveys were repeated in 1962 at Mapperley and Severalls and in 1964 at all three hospitals, exactly two years and four years after the initial investigations. All the patients who were still in hospital were re-interviewed and the same social and behavioural measurements made. Usually, the nurses who completed questionnaires or who gave information to the sociologist about any particular patient were not the same, even for two out of three possible occasions of measurement.

The person who enumerated the patient's personal possessions was different on each occasion. The investigators never looked up their notes of previous interviews or observations at the time of making new ones. Although, in certain instances, it was possible to remember roughly what the previous measurements had been (particularly in the third surveys at Mapperley and Severalls), in general, contamination from this source was likely to be low during the second surveys.

Stated simply, the hypotheses under test in this chapter are that, if environmental conditions improved, symptoms such as flatness of affect, poverty of speech and social withdrawal should also have improved, and clinical change should occur most frequently in those patients whose social milieu had changed. Florid symptoms would be less likely to improve in association with decreased poverty of the social environment, and attitudes to discharge should not, on balance, change at all. Any clinical improvement observed would be independent of changes in the type or dose of phenothiazines prescribed. This chapter is concerned solely with the overall differences found between the two surveys of 1960 and 1964. Changes within each of the three hospitals are compared in Chapter 7, which also presents data collected in 1962 and January 1968.