The annual rate at which patients defaulted from follow up at the Wolverhampton diabetic clinic between 1971 and 1981 was 4-1% overall and 3 5% in white patients. In 1982 a study was started to discover what happened to white patients, born after 1919, who defaulted from the hospital clinic. There were 162 defaulters, of whom 19 had died. Of the remaining 143 patients, 19 were attending another hospital diabetic clinic, 22 had moved out of the area, and 28 refused to participate in the study. Seventy four agreed to participate: 39 were treated with diet, 15 with oral hypoglycaemic agents, and 20 with insulin. They were matched for sex, age, treatment, and duration of diabetes with patients attending the clinic.Non-insulin dependent defaulters were significantly more overweight at diagnosis (40% v 25%; p<0 05) and remained more obese. They developed significantly higher diastolic blood pressure (94 v 86 mm Hg; p<002) and higher haemoglobin A, (HbA,) concentrations (11.7% v 8-4%; p<001). They had significantly more neuropathy at reassessment (15 v 6 out of 54; p<005) and a greater incidence of new retinopathy (p<002), which correlated with their higher diastolic blood pressures (p<0-01) and HbA, concentration (p<0.02). In defaulters who were treated with insulin only the prevalence of neuropathy was significantly different from that in controls (p<005).Defaulters received minimal medical supervision and suffered greater morbidity than regular attenders at the clinic. Introduction Diabetic care in this district comprises a hospital run by consultants and 23 general practice diabetic clinics run independently (mini clinics). Patients who attended either clinic between 1971 and 1981 had a punchcard giving basic information and treatment, as previously described.' Information on attenders is continually updated.In a survey of punchcards carried out in 1982 1117 people attended the hospital clinic and 1061 the mini clinic practices. The overall annual rate of defaulting from the hospital clinic was 4 1% with annual rates of Il% for patients receiving diet alone, 6% for those treated with oral hypoglycaemics, and 4% and 1% for those treated with insulin once or twice a day, respectively. Defaulting from mini clinics is much less of a problem and is not reviewed here.

No abstract

No abstract

No abstract

Summary.In a single blind randomised cross-over study, 40 patients were changed from ordinary bovine to highly purified porcine insulins for a period of 6 months. Half were later rechallenged with bovine insulin. Sequential determinations of IgG insulin binding capacity for bovine insulin were correlated with insulin dose and diabetic control. After changing to highly purified insulins the following correlations were observed between percentage change in insulin dose and change in insulin binding capacity: at 2 months r = 0.35 (p < 0.05), at 4 months r = 0.38 (p < 0.02) and at 6 months r = 0.37 (p < 0.02). When the patients who showed substantial changes in HbA~ were removed from the analysis, the remaining 29 demonstrated a clearer relationship between these two variables (r = 0.56,p < 0.01). Removal of patients with a low initial insulin binding capacity left 18 patients with stable diabetes, and changes in insulin binding capacity and insulin dose showed an even closer correlation for this group (r = 0.77, p < 0.001). A similar degree of positive correlation was observed after rechallenge with bovine insulin. We conclude that the level of circulating insulin antibody affects the dose of insulin required to maintain stable diabetic control.Key words: Insulin, insulin antibody, diabetic control.Most patients treated with ordinary bovine insulins develop circulating insulin antibodies [1]. Higt/ly purified porcine insulins are much less immunogenic and patients changing to these insulins often experience a reduction in insulin requirement [2,3], a fall in circulating insulin binding capacity [4] and an increase in circulating levels of free insulin [5]. Although antibody mediated insulin resistance is unusual [6], injection site lipoatrophy is a common complication of the use of ordinary bovine insulins [7]. Other possible effects of insulin antibodies are less clear. Andersen [8] showed that the presence of antibodies was associated with a reduction in the honeymoon period independent of the level of diabetic control and the hypothesis that insulin antibody might 'buffer' the effect of circulating insulin [9] has not been substantiated. We report a study of the relationship between changes in insulin antibody, insulin dose and diabetic control in established diabetics changing from ordinary bovine soluble and isophane to highly purified porcine equivalents and the effects of rechallenge with the original insulins. Patients and MethodsThe practice of one hospital diabetologist was surveyed to identify all the patients aged between 16 and 65 years who had had diabetes for at least 3 years and had been treated regularly with ordinary bovine soluble and isophane insulins for at least that time. Those who were blind or pregnant or who had other serious medical or psychological illness were excluded. Fifty-nine of the 71 eligible patients consented to be studied and 40 completed the 18 month protocol. These 14 women and 26 men were aged between 16 and 64 years (mean 40 years). The age at diagnosis ranged from 2 to 5...