Muhlack G
Dr. Dana Clarissa Laetsch, born Muhlack, Pharmacist
Division of Clinical Epidemiology and Ageing Research
German Cancer Research Center

Im Neuenheimer Feld 581
D-69120 Heidelberg

Phone: +49 (0)6221 42 1354

2nd Affiliation: Netzwerk AlternsfoRschung, Fellow: Ben Schöttker, PhD

Potentially Inappropriate Medication Use – Risk Factors and Associations with Frailty and Mortality

Summary of PhD thesis

Potentially inappropriate medications are defined as drugs that should be avoided in older adults because of an assumed negative benefit-risk ratio. However, their association with relevant clinical outcomes is unclear. One aim of the dissertation was to conduct a systematic review and meta-analysis of prospective and retrospective cohort studies reporting on the association of potentially inappropriate medication intake with mortality and cardiovascular events. In an own data analysis, I then examined the association between these medications and frailty. Furthermore, I assessed the changes in the prevalence of potentially inappropriate medications over a 6-year period and identified the determinants for their current and future use with a specific focus on geriatric syndromes.

Relevant studies for the systematic review were retrieved from the databases MEDLINE and Web of Science. Meta-analyses were conducted with a random effects model. At first, thirteen studies were included. The association of potentially inappropriate medications with overall mortality was not statistically significant (risk ratio [95% confidence interval]: 1.13 [0.95, 1.35]). However, the majority of studies showed a high risk of specific forms of bias. These biases can be excluded by applying a new user design. It ascertains that adverse events occurring early in therapy are recorded. After restricting the meta-analysis to three studies with a new user design, the association between the use of potentially inappropriate medications and mortality was statistically significant (1.59 [1.45, 1.75]). All three studies examined a version of the Beers criteria. Only one study focused on cardiovascular events and found no statistically significant association. However, the study was not conducted with a new user design. Further new user design studies are required for cardiovascular outcomes and to compare the predictive value of different PIM criteria for mortality.

For my own analyses, the data of the German population-based ESTHER cohort of 2878 adults, aged 60 years and older, with follow-up from October 2008 to September 2016 was used. Potentially inappropriate medications were defined by the PRISCUS list, the EU(7)-PIM list, and the 2015 Beers criteria. Their prevalence was 13.7 %, 37.4 % and 26.4 % at baseline and decreased to 12.3 %, 36.5 %, and 23.1 %, respectively, 6 years later. Unadjusted prevalences in participants with any geriatric syndrome (frailty, comorbidity, functional or cognitive impairment) were approximately twice as high as in robust older adults. Determinants of the use of ≥ 1 potentially inappropriate medication were identified in multivariable logistic regression (cross-sectional analysis) and weighted generalized estimating equation models (longitudinal analysis). Cognitive impairment was statistically significantly associated with the use of potentially inappropriate medication of all three criteria in the cross-sectional (odds ratio point estimates: 1.90 to 2.21) but not in the longitudinal models. In contrast, frailty, comorbidity, and functional impairment were statistically significantly associated with the use of potentially inappropriate medications of at least one of the three criteria in both models. However, the associations varied for the different lists, and in the longitudinal analysis, associations were only statistically significant for medications of the Beers criteria (odds ratios [95% confidence interval]: frailty (2.23 [1.15, 4.31]), comorbidity per 5 total comorbidity score points (1.21 [1.05, 1.38]), and functional impairment (1.51 [1.00, 2.27])). Other statistically significant determinants of the incidence of potentially inappropriate medications (any definition) were female sex, age, coronary heart disease, heart failure, biomarkers of the metabolic syndrome, and history of ulcer, depressive episodes, hip fracture, or any cancer. In summary, older adults with frailty, comorbidity, cognitive impairment, and functional disability had higher odds of taking potentially inappropriate medications or getting a prescription in the future (exception: cognitive impairment).

In propensity score-adjusted logistic and Cox regression models, associations between baseline use of potentially inappropriate medication and prevalent as well as incident frailty were investigated. Frailty was assessed using a modified version of the Fried frailty phenotype. 261 participants were frail at baseline and 423 became frail during 6 years of follow-up. Only the use of drugs that according to the Beers criteria should be avoided in cognitively impaired patients (anticholinergics, benzodiazepines, Z-drugs, and antipsychotics) was statistically significantly associated with prevalent frailty (1.51 [1.04, 2.17]). The strength of the association was comparable for all frailty components. Similarly, in longitudinal analyses, only use of medications of the Beers criteria to avoid in cognitively impaired patients was associated with incident frailty albeit not statistically significantly (Hazard ratio [95% confidence interval]: 1.19 [0.84, 1.68]). Therefore, the association of potentially inappropriate medication use and frailty seems to be restricted to drug classes which can induce frailty symptoms.

Overall, based on the evidence summarized in my systematic review and the findings of my original data analyses, I recommend physicians to avoid potentially inappropriate medications for older adults whenever feasible. Practitioners should be especially cautious when prescribing new drugs for patients with geriatric syndromes or depression, heart failure, coronary heart disease, metabolic syndrome, and history of an ulcer. All clinical conditions were identified as risk factors for (future) use of potentially inappropriate medication in my analyses. Moreover, physicians are advised to perform frailty assessments before and after prescribing anticholinergics, benzodiazepines, Z-drugs, and antipsychotics to older patients and to reconsider treatment decisions in case of negative performance changes. 



  • Muhlack DC, Hoppe LK, Saum KU, Haefeli WE, Brenner H, Schöttker B. Investigation of a possible association of potentially inappropriate medication for older adults and frailty in a prospective cohort study from Germany. Age Ageing 2019;49(1):20-5.
  • Muhlack DC, Hoppe LK, Stock C, Haefeli WE, Brenner H, Schöttker B. The associations of geriatric syndromes and other patient characteristics with the current and future use of potentially inappropriate medications in a large cohort study. Eur J Clin Pharmacol 2018;74(12):1633-1644.
  • Muhlack DC, Hoppe LK, Weberpals J, Brenner H, Schöttker B. Association of potentially inappropriate medication at older age with cardiovascular events and overall mortality: a systematic review and meta-analysis of cohort studies. J Am Med Dir Assoc. 2017;18(3):211-220.



2007-2012 Studies of Pharmacy at the  Carolo-Wilhelmina University in Brunswick
2012-2013 Pharmaceutical Trainee
12/2013 Registration as a Pharmacist
2014-2015 Project collaborator at the consulting service for medications and medication problems at the Departement for Clinical Pharmacology at the Technical University (TU) of Dresden
02/2016-01/2019   Junior Researcher, Network Aging Research (NAR), Heidelberg University
kutsubinas: Email
Latest Revision: 2021-09-21
zum Seitenanfang/up