SummaryThree instruction schemes for self-medication in older patients were designed and compared to see whether they improved drug compliance.

In a prospective study of patients with fractures of the proximal femur from a defined population it was found that age, mental test score, type of residence, place of injury and hospital of admission were all significantly associated with the likelihood of survival to six months. When these factors were taken into account no additional association between outcome and recorded associated disease, grade of surgeon operating, cigarette smoking, type of operation or social class was demonstrable. These findings have implications for the design of controlled trials of different types of treatment for fractures of the proximal femur.

No abstract

No abstract

SUMMARY The effect of counselling by a pharmacist on medication errors was assessed in fifty‐three patients aged 65 years and over attending a day hospital. Despite random allocation to either the counselled or the uncounselled (‘control’) group, patients in the counselled group were making fewer errors than those in the control group even before they received instruction from the pharmacist. There was no evidence that those in the counselled group made fewer errors or complied better with their treatment as a result of counselling.