A novel superior medication-based chronic disease score predicted all-cause mortality in independent geriatric cohorts

J Clin Epidemiol. 2019 Jan:105:112-124. doi: 10.1016/j.jclinepi.2018.09.004. Epub 2018 Sep 22.

Abstract

Objectives: On the basis of current treatment guidelines, we developed and validated a medication-based chronic disease score (medCDS) and tested its association with all-cause mortality of older outpatients.

Study design and setting: Considering the most prevalent chronic diseases in the elderly German population, we compiled a list of evidence-based medicines used to treat these disorders. Based on this list, a score (medCDS) was developed to predict mortality using data of a large longitudinal cohort of older outpatients (training sample; MultiCare Cohort Study). By assessing receiver-operating characteristics (ROC) curves, the performance of medCDS was then confirmed in independent cohorts (ESTHER, KORA-Age) of community-dwelling older patients and compared with already existing medication-based scores and a score using selected anatomical-therapeutic-chemical (ATC) codes.

Results: The final medCDS score had an ROC area under the curve (AUC) of 0.73 (95% CI 0.70-0.76). In the validation cohorts, its ROC AUCs were 0.79 (0.76-0.82, KORA-Age) and 0.74 (0.71-0.78, ESTHER), which were superior to already existing medication-based scores (RxRisk, CDS) and scores based on pharmacological ATC code subgroups (ATC3) or age and sex alone (Age&Sex).

Conclusions: A new medCDS, which is based on actual treatment standards, predicts mortality of older outpatients significantly better than already existing scores.

Keywords: Elderly; Medication-based chronic disease score; Mortality; Multimorbidity; Risk assessment.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Aged
  • Aged, 80 and over
  • Cause of Death
  • Chronic Disease* / epidemiology
  • Chronic Disease* / therapy
  • Female
  • Germany / epidemiology
  • Humans
  • Independent Living / statistics & numerical data*
  • Male
  • Medication Therapy Management / statistics & numerical data*
  • Mortality
  • Multimorbidity
  • Predictive Value of Tests
  • Prognosis
  • ROC Curve
  • Research Design
  • Risk Assessment / methods*