Affiliations 

  • 1 University of Newcastle, University Drive, Callaghan, NSW, 2308, Australia. kaeshaelyathiruchelvam@uon.edu.au
  • 2 University of Newcastle, University Drive, Callaghan, NSW, 2308, Australia
Aging Clin Exp Res, 2021 Jul;33(7):1919-1928.
PMID: 32909246 DOI: 10.1007/s40520-020-01693-y

Abstract

BACKGROUND: Older people use many medications, but combinations of medications used among the oldest old (≥ 80 years) are not commonly reported.

AIMS: This study aimed to determine common combinations of medications used among women aged 77-96 years and to describe characteristics associated with these combinations.

METHODS: A cohort study of older women enroled in the Australian Longitudinal Study on Women's Health over a 15-year period was used to determine combinations of medications using latent class analysis. Multinomial logistic regression was used to determine characteristics associated with these combinations.

RESULTS: The highest medication users during the study were for the cardiovascular (2003: 80.28%; 2017: 85.63%) and nervous (2003: 66.03%; 2017: 75.41%) systems. A 3-class latent model described medication use combinations: class 1: 'Cardiovascular & neurology anatomical group' (27.25%) included participants using medications of the cardiovascular and nervous systems in their later years; class 2: 'Multiple anatomical group' (16.49%) and class 3: 'Antiinfectives & multiple anatomical group' (56.27%). When compared to the reference class (class 1), the risk of participants being in class 3 was slightly higher than being in class 2 if they had > 4 general practitioner visits (RRR 2.37; 95% CI 2.08, 2.71), Department of Veterans Affairs' coverage (RRR 1.59; 95% CI 1.36, 1.86), ≥ 4 chronic diseases (RRR 3.16; 95% CI 2.56, 3.90) and were frail (RRR 1.47; 95% CI 1.27, 1.69).

CONCLUSION: Identification of combinations of medication use may provide opportunities to develop multimorbidity guidelines and target medication reviews, and may help reduce medication load for older individuals.

* Title and MeSH Headings from MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine.