A method of observing and modifying teacher attention to appropriate child responses in preschool classrooms was developed. Two teachers with no formal training in reinforcement principles were observed for a baseline of eight days. Teacher A, who displayed a lower baseline rate of attending to appropriate child responses, was trained first. Teacher B was simply observed during the first part of the training condition for Teacher A. During training, A received feedback which included definitions of appropriate child responses, her frequency of attending to appropriate child responses, her total percentage of attending to appropriate child responses, and her frequency of failing to attend to appropriate child responses. Teacher B was then trained in a similar way. Both teachers showed an increase in attending to appropriate child responses subsequent to the onset of experimental feedback.

This prospective study of 60 stable, HIV-infected children in an economically deprived setting was designed to document anthropometric and micronutrient disturbances. Investigations included CD4+ counts, anthropometry and plasma levels of albumin, transthyretin, retinol-binding protein (RBP), vitamins A, B6, E and B12, and folate, zinc and copper. The median age was 25 months. Thirty-two per cent had mild, 48% moderate and 20% severe clinical features, and 80% were moderately or severely immunosuppressed. Twenty-eight per cent had a weight Z-score <-2.0 and 58% a height Z-score <-2.0. Many children had micronutrient deficiencies: albumin (70%), transthyretin (100%), RBP (85%), vitamins A (80%), B6 (37%), E (37%) and B12 (5%), zinc (20%) and copper (25%). Sixty-two per cent had two or more trace element or vitamin deficiencies. There was a weak association between micronutrient status and disease status. Micronutrient concentrations did not correlate with chronological age, height-for-age or weight-for-age. CRP was elevated in 53% but did not correlate with any of the micronutrient concentrations. Micronutrient deficiencies were more common and micronutrient concentrations lower in children over 24 months of age.

The spectrum of PID in South Africa was similar to international trends. The declining mean age of diagnosis indicated improved recognition of PID. Future research should focus on identifying children with PID more effectively.

Patients with a history of rheumatic fever had statistically increased expression of the D8/17 marker. However, the actual percentages in this observational study were markedly lower than in other populations, ranging from 0.14%-1.53% compared to 11.6%-39.3%. The D8/17 marker would be an impractical screening tool in the South African population.