This study aimed to describe medicine use in very old age in different social classes, taking into account differences in self-reported health. A cross-sectional interview study design was used. The study population was made up of a nationwide representative sample of the very old in Sweden and included all people aged ge; 75 years, earlier interviewed in the Swedish Level of Living surveys. Study measurements were drug use, self-reported health (i.e. symptomssol;diseases), and social class. The response rate was 95.4percnt; (n equals; 537, age range 77 to 98 years). Women accounted for 61percnt; of the total sample; they used more drugs than men mainly because of the large number of male nonusers. No differences could be found with respect to the number of drugs used in different age groups, nor in different social classes. There were some differences in types of drugs used between social classes. When adding symptomssol;diseases into a logistic regression model, some significance differences disappeared or became less pronounced.In conclusion, it was considered necessary to go beyond tabulating the number of drugs or frequencies of medication in order to obtain a complete picture of drug use by older adults. One must consider categories of drugs and types of health conditions. A global health measure was not very useful in predicting drug use, while specific symptoms were related in expected ways. A limitation of the study was reliance on self-reported health and drug use; however, questions about specific symptoms (e.g. pain in shoulder) are less affected by subjective bias, and self-reported drug use is not as much an overestimation of drug use as alternative methods. The results showed that differences between gender, age and social class subgroups could be quite subtle. For social class, differences only emerged when health was controlled for. Finally, age differences in drug use were incompletely accounted for by self-reported health.
展开▼
