Numbers (percentages) of patients complying with and defaultingfrom treatment for tuberculosis in Botswana, 1984-90 No(") No(%) Year complying

This cross-sectional study collected baseline data on the health behaviours of a large population of survivors of childhood cancer in the UK, aged 18 -30 years, compared with those of sex-and age-matched controls. Data from 178 young adult survivors of childhood cancer, diagnosed and treated at Bristol Children's Hospital, 184 peers from the survivors' GP practices and 67 siblings were collected by postal questionnaire. Conditional logistic regression analysis showed that, for matched sets of survivors and controls, survivors of a variety of childhood cancers reported lower levels of alcohol consumption (P=0.005), lower levels of cigarette smoking (P=0.027) and lower levels of recreational drug use (P=0.001) than controls. Analysis of matched sets of survivors and siblings showed similar trends but no significant differences. A health behaviour index for each participant was constructed from the data collected on five key health behaviours which influence future health status. Comparison of the means for each case group showed that survivors of childhood cancer were leading healthier lives than controls or siblings. This finding was expressed most clearly as the difference in the means of the health behaviour index for each case group, derived from five health behaviours (one-way ANOVA, P50.001). British Journal of Cancer (2002) It is only in the last 30 years that advances in the treatment of childhood cancer have resulted in the majority of patients surviving into adulthood, with the probability of being able to live a normal life-span. Recent research (Coleman et al, 1999, cited in CRC, 1999 shows that several childhood cancers now have survival rates in excess of 75%. However, an increased risk of early death from causes other than the original cancer for survivors, when compared with their siblings, has been found (Nicholson et al, 1994). There is also considerable evidence to show that survivors of childhood cancer are at increased risk of developing other late-effects including a second cancer. These may be due to a combination of genetic factors (Li et al, 1988) and/or the consequences of damage to various organs caused by the original treatment (Hawkins and Stevens, 1996).It is well known that lifestyle factors, including smoking, consumption of alcohol, diet, sexual and sun behaviour, influence the risk of developing cancers and other health conditions in the general population. Since survivors of childhood cancer have been shown to be a population susceptible to further ill-health, it is likely that non-avoidance of known 'risk' health behaviours by this vulnerable population of young people may further increase this risk. Until recently, little research has been carried out on lifestyles of survivors, but with increasing numbers reaching adulthood their adjustment and assimilation into the community and their health behaviours have become a focus for research. Do the health behaviours and lifestyles of survivors of childhood cancer differ from those of their peers, and if so, how? No British data i...

Four examples are reported of an unusual noncystic intraosseous lesion which does not conform to any hitherto recognized entity and which can be mistaken, not only by the general histopathologist but by the osteoarticular pathologist, for a variety of other conditions, including sarcoma and giant‐cell tumor. They were in patients aged 5 to 13 years; three in the spine, one in the ethmoid. Local excision, supplemented by low‐dose radiotherapy in cases with cord compression, produced a satisfactory outcome in all cases. At presentation the radiologic findings were nonspecific but, following treatment, an eggshell rim of bone developed in those lesions which had been incompletely excised. Histologically, they are characterised by florid fibroblastic or fibrohistiocytic proliferation, osteoblastic differentiation with osteoid production, areas rich in osteoclast‐type giant cells, aneurysmal sinusoids, and occasional foci of degenerate calcifying fibromyxoid tissue. Because this combination of histologic features can be found in the solid parts of aneurysmal bone cyst and in no other condition, at this centre we have regarded this lesion as a variant of aneurysmal bone cyst devoid of any cystic component.