Abstract
Background
Listing tools were found to ameliorate drug treatment in older people; the FORTA (Fit-fOR-The-Aged) list is a clinically validated positive-negative list of medication appropriateness. Here, we retrospectively analyze longitudinal correlations between the FORTA score and key measures of physical and cognitive function in older people.
Methods
504 participants of a multi-center cohort study (AgeCoDe/AgeQualiDe) for whom the FORTA score (sum of over- and under-treatment errors) had been assessed were studied at three follow-up (FU) time points (FU 6–8; mean age range 87.9–89.7 years); comparisons between data at these FUs separated by 10 months were available for 292–328 patients.
Results
The univariate analysis of the association between FORTA_Delta_76 (change of FORTA score between FU 6 and 7) and ADL (Activities of Daily Living)_Delta_76 (− 0.155, p < 0.01) and between FORTA_Delta_76 and MMSE (Mini-Mental State Examination)_Delta_76 (− 0.203, p < 0.01) revealed significant correlations. Multivariable analysis (using a forward selection model, p < 0.05) revealed a significant association between FORTA_Delta_76 and MMSE_Delta_76 (p < 0.05). Univariate analyses for other comparisons were only significant for FORTA_Delta_86 and MMSE_Delta_86.
Conclusion
This study indicates that longitudinal non-interventional changes of the FORTA score as an integral index of medication appropriateness are associated with changes in ADL and MMSE: the lower this score the better the functional outcome. These findings are in line with earlier interventional data and underscore the potential of FORTA to improve clinical endpoints in older people.
Supplementary Information
The online version contains supplementary material available at 10.1007/s40266-026-01301-0.
Key Points
| This study examines the longitudinal associations between the appropriateness of drug treatment and key geriatric outcomes in older people. |
| Relevant clinical outcomes are significantly associated with longitudinal non-interventional alterations of the appropriateness of drug treatment as measured by the FORTA score. |
Introduction
Polypharmacy, as defined by the use of five or more drugs, as a consequence of multimorbidity in older adults [1–6] frequently results in avoidable loss of functional independence, lowered physical performance, deterioration in mental abilities, hospitalization and even death [2, 6–11]. Lists of drugs and/or action points have been developed to address this problem, such as the Beers Criteria®. Only a few of these are based on patients’ characteristics, such as diagnoses, severity, functionality and patient’s wishes/needs [4, 6]. Such lists may be more appropriate to assist clinicians in tackling the problem of polypharmacy, and to address both under- and over-treatment [6]. Most lists such as the Beers Criteria® [12] present negative lists of drugs to be avoided in older adults [4, 6]. In contrast, the Screening Tool to Alert Doctors to the Right Treatment (START)/Screening Tool of Older Persons’ Prescriptions (STOPP) criteria [13] or the Fit fOR The Aged (FORTA) list [14] include positive and negative action points or labeling of drug treatments [4, 6]. Based on thorough knowledge of patients’ diseases, their severity and on physical and mental abilities, these tools cannot be used by just looking at the drug list of a given patient [4, 6]. Demonstration of beneficial clinical endpoint effects has been largely restricted to those positive-negative lists. The term ‘patient-in-focus listing approach’ (PILA) was proposed as preferable to ‘drug-oriented listing approach’ (DOLA) [4, 6] to underline this overall superiority. For example, in a randomized controlled trial (VALFORTA) in older hospitalized patients, FORTA has been shown to significantly improve medication quality as determined by the FORTA score. In addition, important clinical outcomes such as adverse drug reactions or activities of daily living (ADL) were also improved [6, 15].
The quantitative assessment of the appropriateness of drug treatment according to FORTA results in the FORTA score representing the sum of over- and under-treatment errors [6]. Associations between the FORTA score and ADL, IADL (instrumental activities of daily living), MMSE (Mini-Mental State Examination), sleep quality and other parameters have already been described [16, 17] in addition to the interventional effects of the FORTA list. While these were cross-sectional association studies (e.g. assessing the FORTA score and clinical outcomes for one timepoint), here we report on the longitudinal associations of these parameters over time; data from three study visits covering an observation time of 30 months were compared for patients in the AgeCoDe study.
Methods
Study Design and Study Population
This work was based on the Study on Ageing, Cognition and Dementia (AgeCoDe), a multi-center (Bonn, Düsseldorf, Hamburg, Leipzig, Mannheim and Munich), population-based, longitudinal cohort study in primary care patients commencing in 2003/2004 [6].
Primary care patients aged 75 years and above without the diagnosis of dementia at baseline were recruited through general practitioners’ (GP) offices. Follow-up (FU) assessments were performed at 1.5-year intervals on average [6, 18] until FU 7; from FU 7–8 the interval was only 10 months. The 8th follow-up finished in 2016 [6, 19].
Data from FU 7 onwards were collected from AgeQualiDe (a study on needs, health service use, costs and health-related quality of life in a large sample of oldest-old primary care patients [85+]). AgeQualiDe represents an extension of the AgeCoDe study, continuing largely comparable data collection procedures, diagnostic criteria and core outcome assessments to enable longitudinal analyses across study waves. However, some modifications were introduced, including shorter follow-up intervals as indicated above and additional measures (e.g., quality of life and health care utilization), meaning the methodologies were expanded.
Attrition throughout the study up to FU 6 is described elsewhere [18]. At FU 3, there were 1161 participants, 69 died and 145 dropped out, at FU 4 these numbers were 977, 76, 82; at FU 5 they were 822, 74, 66; at FU 6 they were 652, 62, 68. Similarly, for the following FUs in AgeQualiDe it is stated that attrition was mainly due to death and dropout [20, 21]. In detail, at FU 7, 136 participants died and 46 refused participation; at FU 8, 78 died and 17 refused; and at FU 9, 92 died and 18 refused [20, 21].
The main outcomes of this retrospective analysis of longitudinal cohort data were to find multivariate and univariate associations between the FORTA score and clinical variables, and statistical comparisons of these variables for patients showing an increase and those showing a decrease of the FORTA score; thus, the cutoff for this dichotomization of the pre/post FORTA score difference (delta FORTA score) was set to zero.
Data Collection
Relevant data collected at the 6th through 8th FU (FU 6–8) “included drug use (ATC codes), age, gender, GP diagnoses and blood pressure. Data were electronically entered into the database” [6]. At FU 6–8 of the AgeCoDe/AgeQualiDe study, patients (by patient ID) were included if data sufficient for comparison were present for FU 7–6/8–7/8–6 [6].
FORTA (Fit fOR The Aged) Diagnoses
Data were not available for all FORTA diagnoses. The alignments of non-FORTA diagnoses have been detailed before [16]; in brief, any of the following diagnoses were assumed to reflect the FORTA diagnosis ‘gastrointestinal disease’: gastritis, reflux gastritis, reflux, esophageal carcinoma and gastrointestinal bleeding. Similarly, stroke, cerebral infarction, stenosis of the afferent cerebral arteries and transient ischemic attacks were attributed to the FORTA diagnosis ‘stroke’ [6].
Further details of these studies are provided elsewhere [18, 19, 22–30] and in the earlier papers on the FORTA analysis of those data [6, 15].
Ethics
Approval from the ethics committees of all the participating centers [23] had been obtained for the AgeCoDe and AgeQualiDe studies, which were conducted in compliance with the Code of Ethics of the World Medical Association [6, 31]. Furthermore, ethics approvals that were previously received (reference number: 2007-253E-MA) and recently renewed by the ethics committee in Mannheim, University of Heidelberg (reference number: TEMP558252-AF 11) cover the work in this paper [6].
Determination of the FORTA Score
In brief, “the FORTA-list assigns 4 FORTA classes to drugs that are defined as follows” [6]:
“Class A (A-bsolutely) = indispensable drug, clear-cut benefit in terms of efficacy/safety ratio proven in elderly patients for a given indication
Class B (B-eneficial) = drugs with proven or obvious efficacy in the elderly, but limited extent of effect or safety concerns
Class C (C-areful) = drugs with questionable efficacy/safety profiles in the elderly, to be avoided or omitted in the presence of too many drugs, lack of benefits or emerging side effects; review/find alternatives
Class D (D-on’t) = avoid in the elderly, omit first, review/find alternatives” [6].
“The related FORTA-score is the sum of medication errors classified as over- and/or under-treatment errors in an individual patient as checked against these labels. An error was counted if an indication was not appropriately treated though beneficial options (FORTA A or B exist – under-treatment) or if a prescription was suboptimal regarding the FORTA categories (e.g., FORTA C though A or B drugs exist) or not indicated (over-treatment)” [6].
“Proton-pump-inhibitors are often not indicated and would trigger an overtreatment error; oral anticoagulation is strictly indicated in atrial fibrillation, and the absence of a positively labeled oral anticoagulant (e.g., apixaban) would be considered as undertreatment error. To apply FORTA, a demand analysis for drug treatment is required and the determination of the FORTA score relies on it.” [6].
A more detailed description of the FORTA score can be found elsewhere [14, 32] and in the first paper on FORTA in AgeCode [6].
For FU 6, the FORTA scores have already been published by Pazan et al. [6, 16]. The score was determined at FU 7 and 8 in the same way.
Statistical Analysis and Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement
A preceding sample size analysis [16] showed that 310 patients from this cohort would be sufficient to detect a minimal clinically important difference (MCID, one score point) assuming a standard deviation of 2.7 (power = 0.9). The MCID for FORTA changes was oriented on the difference of FORTA scores in the interventional VALFORTA trial [15], which was 1.7, and resulted in the clinical benefits; a clear dose–response relationship could not be established from these data, thus, the assumption of 1 as MCID is consensual, but within the frame of clinical success.
Associations between changes in FORTA score (delta FORTA) and corresponding changes in ADL, IADL and MMSE were analyzed for three follow-up pairings (FU7 vs FU6, FU8 vs FU7 and FU8 vs FU6) using Spearman’s correlation analysis, with all variables treated as continuous variables. Besides, to minimize the risk of type I error, the Bonferroni correction was applied reflecting the total number of tests. Changes in FORTA score (delta FORTA) were analyzed as a continuous variable in order to capture the full range and magnitude of change and to avoid loss of information associated with dichotomization. Associations between FORTA score changes and clinical outcomes (MMSE, ADL and IADL) were assessed using multivariable linear regression models, adjusting for potential confounders including age, sex, number of medications and number of comorbidities. Clinically relevant decline was defined a priori as a decrease of ≥ 1 point in MMSE and IADL, and ≥ 5 points in ADL, reflecting differences in scale granularity and established clinical interpretability. Receiver operating characteristic (ROC) curve analyses were conducted to evaluate the ability of continuous FORTA score changes to discriminate between participants with and without clinically relevant decline. The area under the curve (AUC) was used as a measure of discriminative performance. Optimal cut-offs were estimated using the Youden index and are reported for exploratory purposes only, given the limited discriminatory ability observed.
To facilitate comparability with previous approaches, an additional sensitivity analysis was performed in which participants were categorized according to the direction of FORTA score change (delta FORTA > 0 vs < 0). Group differences were assessed using the Wilcoxon rank-sum test; however, this dichotomization was considered secondary due to its inherent limitations. The strength of correlation was classified according to [33]; all correlations reported here are ‘weak’ (e.g., < ± 0.3).
Statistical significance was assumed at p < 0.05. Statistical analyses were performed using SAS Version 9.4 software for Windows (SAS Institute Inc., Cary, NC, USA) and R version 4.5.1 (R Foundation for Statistical Computing, Vienna, Austria). Moreover, the STROBE checklist (Electronic Supplementary Material 1 [ESM 1] [34]) was used to ensure the inclusion of relevant data.
Results
The demographics, clinical characteristics and numbers of patients included in FUs 6–8 are shown in Table 1.
Table 1.
The demographics and clinical characteristics of the cohort at FU 6 (total no. = 504), FU 7 (total no. = 430) and FU 8 (total no. = 433)
| Items | FU 6 | FU 7 | FU 8 |
|---|---|---|---|
| Mean FORTA score (median; range) | 6.2 (6; 0–18) | 5.1 (5; 0–15) | 5.2 (5; 0–15) |
| Mean age (median; range) | 87.9 (87.5; 83–101) | 88.9 (88.0; 82–99) | 89.7 (89.0; 80–99) |
| Mean number of diseases (median; range) | 5.3 (5; 0–14) | 4.8 (5; 1–11) | 4.9 (5; 1–14) |
| Mean number of medications (median; range) | 7.99 (7; 5–23) | 8.1 (7; 5–22) | 8.1 (7; 5–21) |
| Female gender, % (n) | 69.6 (351) | 69.5 (299) | 71.6 (310) |
| Hypertension, % (n) | 82.3 (415) | 80.7 (347) | 80.6 (349) |
| Arthritis, % (n) | 61.1 (308) | 63.3 (272) | 66.3 (287) |
| Lipid metabolism disorder, % (n) | 57.3 (289) | 54.2 (233) | 52.4 (227) |
| Coronary heart disease, % (n) | 41.8 (211) | 40.2 (173) | 37.9 (164) |
| Heart failure, % (n) | 36.7 (185) | 33.7 (145) | 37.6 (163) |
| Cardiac arrhythmias, % (n) | 35.3 (178) | 34.9 (150) | 36.95 (160) |
| Type II diabetes mellitus, % (n) | 29.9 (151) | 29.8 (128) | 28.9 (125) |
| Depression, % (n) | 28.9 (146) | 25.8 (111) | 24.7 (107) |
| Dementia, % (n) | 17.8 (90) | 19.5 (84) | 23.1 (100) |
Values for FU 6 have already been published [6]
FORTA Fit fOR The Aged, FU follow-up, N number of cases
FU6-FU7
A total of 316 patients with complete data available for both FU 6 and FU 7 could be included. The univariate analysis of the associations between FORTA_Delta_76 (change of the FORTA score between FU 6 and 7, analogous for all abbreviations) and and ADL_Delta_76 (− 0.155*, p < 0.01) and between FORTA_Delta_76 and MMSE_Delta_76 (− 0.203*, p < 0.01) revealed significant correlations. The association between FORTA_Delta_76 and IADL_Delta_76 was, however, not significant (− 0.068*, p > 0.05).
Significance was even observed in the more stringent analysis using the Bonferroni adjustment (i.e., after decreasing the level of significance to α/3 = 0.0166). The correlation for FORTA_Delta_76 and IADL_Delta_76 was not significant (− 0.067*, p > 0.05). In addition, we compared the changes in results of the physical function tests (ADL_Delta_76 and IADL_Delta_76) and cognitive test (MMSE_Delta_76) between the patients with a Delta FORTA score of 0 or more and the patients with a Delta FORTA score of − 1 or lower (Fig. 1).
Fig. 1.
For a cutoff of 0 of the change in the FORTA score between FU 6 and FU 7, ADL (A) and MMSE (C) were significantly higher in those patients with the lower FORTA scores; the difference for IADL (B) was not significant (Wilcoxon rank sum test). ADL activities of daily living, FORTA Fit fOR The Aged, FU follow-up, IADL instrumental activities of daily living, MMSE Mini-Mental State Examination
ROC analyses indicated only limited discriminatory ability of FORTA score changes for clinically relevant decline (AUC 0.54–0.62). These findings suggest that FORTA score changes do not operate as a threshold-based predictor, but rather reflect a continuous and gradual association with functional and cognitive outcomes. Corresponding ROC curves are shown in the ESM.
The multivariable analysis using a forward selection linear regression model revealed a significant association between FORTA_Delta_76 and MMSE_Delta_76 (β = − 0.123, SE = 0.061, t = − 2.02, p = 0.045) adjusting for the following confounders: gender, age, number of medications (FU 6 and 7) and number of diseases (FU 6 and 7). No significant associations were found for ADL_Delta_76 (β = − 0.287, SE = 0.329, t = − 0.87, p = 0.38) and for IADL_Delta_76 (β = − 0.048, SE = 0.033, t = − 1.47, p = 0.14) in the multivariable analysis.
FU7–FU8
In the analysis of FU 8 compared with FU 7, a total of 328 patients were included based on the availability of data. Except for the average FORTA score, the other average scores (ADL, IADL, MMSE) showed a decrease at the FU 8 examinations compared with the FU 7 examinations. No statistically significant correlations (p > 0.05) were found between FORTA_Delta_87 and the corresponding deltas of ADL, IADL, or MMSE (ADL_Delta_87, IADL_Delta_87, or MMSE_Delta_87; correlation coefficients were − 0.068, − 0.012 and − 0.113, respectively). In addition, we compared the changes in results of the physical function tests (ADL_Delta_87 and IADL_Delta_87) and cognitive test (MMSE_Delta_87) between the patients with a Delta FORTA score of 0 or more and the patients with a Delta FORTA score of − 1 or lower. These comparisons were not significant (Supplementary Fig. 1 in the ESM).
Upon examining the impact of confounding factors, we identified a significant positive association between FORTA_Delta_87 and the number of medications at FU 8 (0.115, p < 0.05). For ADL_Delta_87, we observed significant negative correlations with the number of medications at both FU 8 and 7 (− 0.151, p < 0.01 and − 0.136, p < 0.05, respectively), as well as with the number of diseases at FU 8 (− 0.116, p < 0.05). Regarding IADL_Delta_87, significant negative correlations were found with the number of diseases at FU 8 and 7 (− 0.233, p < 0.001 and − 0.122, p < 0.05), as well as with the number of medications at FU 8 and 7 (− 0.203, p < 0.001 and − 0.187, p < 0.001, respectively). However, no significant correlations were detected for MMSE_Delta_87.
FU6–FU8
A total of 292 patients had complete data for both FU 6 and FU 8. The univariate analysis of the association between FORTA_Delta_86 and ADL_Delta_86 and between FORTA_Delta_86 and IADL_Delta_86 did not show significant correlations (p > 0.05; correlation coefficients were − 0.091 and − 0.026, respectively). Yet, there was a significant correlation between FORTA_Delta_86 and MMSE_Delta_86 (− 0.136, p < 0.05); however, with Bonferroni’s adjustment, this significance was lost. In addition, we compared the changes in results of the physical function tests (ADL_Delta_86 and IADL_Delta_86) and cognitive test (MMSE_Delta_86) between the patients with a Delta FORTA score of 0 or more and the patients with a Delta FORTA score of − 1 or lower. These comparisons were not significant (Supplementary Fig. 2, ESM).
Next, we considered several potential confounding factors in our analysis. The analysis of FORTA_Delta_86 did not reveal statistically significant results for any of the confounding factors (Supplementary Table 1, ESM). However, when examining the association between ADL_Delta_86 and the confounding factors, we found a significant negative correlation with patient age, the number of diseases at FU 8 and the number of medications at FU 6 and 8 (Supplementary Table 2, ESM). In the case of IADL_Delta_86, a significant negative correlation was observed only with the number of diseases at FU 8 (− 0.116, p < 0.05). MMS_Delta_86 showed a significant negative correlation with the confounding factor of age (− 0.192, p < 0.01).
Discussion
This retrospective, longitudinal, non-interventional study indicates that changes of the FORTA score are associated with changes in ADL and MMSE: the lower this score, the better the functional outcome. In particular, MMSE as a clinical test for cognitive function was associated with the FORTA score even in a multivariate analysis. These findings are in line with earlier interventional data; they underscore the potential of FORTA-driven medication optimization to improve clinical endpoints in older people.
Improvement of drug treatment in older, multimorbid patients is urgently needed; a fashionable method for amelioration may seem to be de-prescribing, thereby removing ‘bad’ drugs (PIM, potentially inappropriate medications). Unfortunately, this approach has been disappointing in that no PIM were removed or clinical endpoints improved [4, 35–37], including with the American Beers Criteria [36] and the German PRISCUS list [38]. In contrast, two-sided approaches addressing both over- and under-treatment were almost the only ones to be clinically effective [4, 15, 39]; yet, two more recent large interventional trials on the PILA START/STOPP did not demonstrate beneficial clinical outcomes: OPERAM and SENATOR [40, 41]. Even in these PILA trials, implementing the recommendations into practice was rarely achieved.
So far, the FORTA list has been successful in controlled, randomized clinical trials: the FORTA intervention significantly improved the following clinically important endpoints: falls, Barthel index (ADL) and adverse drug reactions, with the latter at a number-needed-to-treat of only 5 [15, 42]. Several association studies were in line with these interventional studies, although not proving causality at all; looking at AgeCoDe data in an earlier paper showed a significant correlation between the FORTA score (see above) with ADL/IADL at baseline and at all FU visits. This association was significant for ADL/IADL even in the multivariable analysis. In addition, the mean FORTA scores were significantly higher in patients with than in those without dementia. Mortality was significantly increased at higher FORTA scores [16]. An association study from the interventional VALFORTA trial [15] revealed significant correlations, though no causality, between the FORTA score and IADL, the Tinetti test, the Essen Questionnaire on Age and Sleepiness, the MMSE and handgrip strength in the univariate analysis, and with IADL, the Tinetti test and the Essen Questionnaire on Age and Sleepiness in the multivariable analysis [17]. However, a retrospective study found no significant association between the FORTA score and outcomes such as MMSE in patients after hospital discharge over a 12-month follow-up period [43]. Another association study by Chen et al. [44] showed that higher FORTA scores were significantly associated with worse ≤ 5-year overall and colorectal cancer-specific survival among older colorectal cancer patients. Besides, in a recent population-based cohort study of over 50,000 community-dwelling older adults, FORTA-defined inappropriate prescribing was strongly associated with geriatric conditions, particularly neurocognitive disorders, depression, Parkinson’s disease, osteoporosis and COPD [45].
All in all, these association studies were mainly cross-sectional analyses at a given point in time/FU. This study indicates that some of these significant correlations may prevail in longitudinal studies, namely for ADL and MMSE, and that longer time periods may be needed for larger or significant changes to occur. In this retrospective longitudinal study, only changes of the FORTA score over time were correlated with corresponding changes of the clinical endpoints. Most significances were seen for the longer interval (18 months) from FU visit 6 to 7; these significances were lost from FU visit 7 to 8, but this interval was considerably shorter (10 months). This may be speculatively interpreted to show the time dependence of clinical effects possibly driven by changes in the FORTA score; these effects may require > 10 months to develop. The differences, however, may also be due to variations in sample size, regression to the mean, increased variability over longer follow-up, reduced sensitivity of change scores, selective dropout of more frail participants, or loss of power rather than a true biological time effect.
It is essential to note that Bonferroni’s correction for multiple testing expectedly reduced the number of significances, indicating that some differences were significant due to confounders only.
Though still non-interventional, this type of analysis can detect associations of clinical effects with changes of the FORTA score rather than of the one-time status of the FORTA score. That means that whatever may have been behind the improvement or deterioration of appropriateness of medication use (measured by the FORTA score) was associated with clinical changes. These findings indicate that attempts to improve medications according to a post-hoc assignment of FORTA errors may be associated with improved outcomes even if not systematically implemented in an interventional trial. In contrast, deteriorating appropriateness of medication use (an increase of the FORTA score) is associated with harm to the patient even if assessed in this non-interventional trial. In any case, this hypothesis-generating association needs to be corroborated in prospective controlled clinical trials.
Despite limited discriminatory performance in ROC analyses, FORTA score changes appear to reflect a gradual, continuous association with functional and cognitive outcomes rather than a distinct threshold effect.
Limitations
The analysis relied on previously collected data from a completed study; thus, information on patient data and diagnoses essential for FORTA evaluations were partly missing [6]. It was necessary to assign some FORTA diagnoses to similar diagnoses that were available [6]. The alignment of diagnoses could have been biased and, thus, influenced the results [6]. Therefore, this validation is restricted to the available diagnoses and, thus, may not be valid for patients in whom all FORTA diagnoses had been prospectively assessed [6].
Similarly, the medication assessment may have been hampered by incomplete data which could not be replenished as FUs were already completed [6].
The inclusion of cases from a larger cohort critically depends on the availability of data essential for this secondary analysis. The availability of data may have introduced further bias [6].
Time intervals were different between FU visits; this could have implications for the size of clinical effects. In addition, repeated measurements from the same participants across follow-ups were analyzed as independent observations. This approach does not account for within-subject correlations and may have influenced variance estimates, though a Bonferroni was applied.
These limitations may also account for the heterogeneity of results if comparing FU 6/7, 7/8 and 6/8, especially due to a shorter interval of FU 7/8, resulting in a smaller number of significant differences for the latter two; on the other hand, for FU 6/7 the stricter analysis by Bonferroni’s correction still showed relevant significances. Another limitation is that our multivariable analyses adjusted only for gender, age, number of medications and number of diseases. Important factors such as frailty trajectory, disease severity and acute medical events were not incorporated, which may have impacted the observed associations. Future research, preferably in prospective studies including these confounders, would help to provide more accurate estimates of the relationships between FORTA scores and post-discharge endpoints. Finally, the results of this study are limited to correlations between the quality of drug treatment and relevant clinical outcomes. As this study was non-interventional, these associations do not imply causality at all and related hypotheses need to be corroborated in interventional, randomized, controlled trials.
Conclusion
These findings underscore that changes in appropriateness of medication use (measured by the FORTA score) according to a post-hoc assignment of FORTA errors may clinically affect patients even if they are not systematically implemented by an interventional trial; here, they reflect trial-independent alterations by usual care. This holds true for both improvements of appropriateness of medication use associated with better clinical performance and deteriorations of appropriateness of medication use associated with worse clinical outcomes.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
The authors thank all participants of the AgeCoDe/AgeQualiDe study for their valuable contribution to this work.
Funding
Open Access funding enabled and organized by Projekt DEAL. No specific grant from any funding agency in the public, commercial, or not-for-profit sectors supported this research. The AgeCoDe study was funded by the German Federal Ministry of Education and Research (Bundesministerium für Bildung und Forschung/BMBF).
Declarations
Conflict of Interest
M. Wehling was employed by AstraZeneca R&D, Mölndal, as director of discovery medicine (translational medicine) from 2003 to 2006, while on sabbatical leave from his professorship at the University of Heidelberg. Since returning to this position in January 2007, he has received lecturing and consulting fees from ASPEN, Bristol Myers, Bayer, Berlin-Chemie, LEO, Mundipharma, Novartis, Pfizer, Polyphor, Helsinn, Allecra, Santhera, Allergan, Novo-Nordisk, Lilly, AstraZeneca, Roche, Heel, Sanofi-Aventis, Shire and Daichii-Sankyo. All other authors declared that they have no conflicts of interest.
Data Availability
Upon submission of a written request and evaluation by the study group, the data used in this study will be provided to other researchers.
Author Contributions
Conceptualization: Martin Wehling, Farhad Pazan. Data curation: Martin Wehling, Johanna Knorr, Christel Weiss, Kathrin Heser, Alexander Pabst, Melanie Luppa, Horst Bickel, Siegfried Weyerer, Michael Pentzek, Hans-Helmut König, Dagmar Lühmann, Carolin van der Leeden, Martin Scherer, Steffi G. Riedel-Heller, Michael Wagner, Farhad Pazan. Formal analysis: Farhad Pazan, Martin Wehling, Johanna Knorr, Christel Weiss. Funding acquisition: Melanie Luppa, Michael Pentzek, Horst Bickel, Siegfried Weyerer, Hans-Helmut König, Martin Scherer, Steffi Riedel-Heller, Michael Wagner. Investigation: Martin Wehling, Johanna Knorr, Christel Weiss, Kathrin Heser, Alexander Pabst, Melanie Luppa, Horst Bickel, Siegfried Weyerer, Michael Pentzek, Hans-Helmut König, Dagmar Lühmann, Carolin van der Leeden, Martin Scherer, Steffi G. Riedel-Heller, Michael Wagner, Farhad Pazan. Methodology: Farhad Pazan, Johanna Knorr, Christel Weiss, Martin Wehling. Project administration: Farhad Pazan, Martin Wehling. Resources: Martin Wehling, Johanna Knorr, Christel Weiss, Kathrin Heser, Alexander Pabst, Melanie Luppa, Horst Bickel, Siegfried Weyerer, Michael Pentzek, Hans-Helmut König, Dagmar Lühmann, Carolin van der Leeden, Martin Scherer, Steffi G. Riedel-Heller, Michael Wagner, Farhad Pazan. Software: OptiMedis. Supervision: Farhad Pazan, Martin Wehling. Validation: N/A. Visualization: Farhad Pazan, Johanna Knorr, Martin Wehling. Writing—original draft: Martin Wehling, Johanna Knorr, Christel Weiss, Kathrin Heser, Alexander Pabst, Melanie Luppa, Horst Bickel, Siegfried Weyerer, Michael Pentzek, Hans-Helmut König, Dagmar Lühmann, Carolin van der Leeden, Martin Scherer, Steffi G. Riedel-Heller, Michael Wagner, Farhad Pazan. Writing—review and editing: Martin Wehling, Johanna Knorr, Christel Weiss, Kathrin Heser, Alexander Pabst, Melanie Luppa, Horst Bickel, Siegfried Weyerer, Michael Pentzek, Hans-Helmut König, Dagmar Lühmann, Carolin van der Leeden, Martin Scherer, Steffi G. Riedel-Heller, Michael Wagner, Farhad Pazan.
Ethics Approval
Approval from the ethics committees of all the participating centers had been obtained for the AgeCoDe and AgeQualiDe studies, which were conducted in compliance with The Code of Ethics of the World Medical Association. Furthermore, ethics approvals which were previously received (reference number: 2007-253E-MA) and recently renewed by the ethics committee in Mannheim, University of Heidelberg (reference number: TEMP558252-AF 11) cover the work in this paper.
Informed Consent/Statement of Human Rights
Informed consent was obtained from all individual participants included in the study.
Consent for Publication
Not applicable.
Code Availability
Not available.
Clinical Trial Number
Not applicable.
Footnotes
Michael Wagner and Martin Wehling were shared last authorship.
References
- 1.United Nations, D.o.E.a.S.A., Population Division (2020). World Population Ageing 2019 (ST/ESA/SER.A/444). 2019. [02 Nov 2020]. https://www.un.org/en/development/desa/population/publications/pdf/ageing/WorldPopulationAgeing2019-Report.pdf.
- 2.Mc Namara KP, et al. Health professional perspectives on the management of multimorbidity and polypharmacy for older patients in Australia. Age Ageing. 2017;46(2):291–9. [DOI] [PubMed] [Google Scholar]
- 3.Molokhia M, Majeed A. Current and future perspectives on the management of polypharmacy. BMC Fam Pract. 2017;18(1):70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Pazan F, Kather J, Wehling M. A systematic review and novel classification of listing tools to improve medication in older people. Eur J Clin Pharmacol. 2019;75(5):619–25. [DOI] [PubMed] [Google Scholar]
- 5.Gutiérrez-Valencia M, Martínez-Velilla N, Vilches-Moraga A. Polypharmacy in older people: time to take action. Eur Geriatr Med. 2019;10(1):1–3. [DOI] [PubMed] [Google Scholar]
- 6.Wehling M, Weindrich J, Weiss C, Heser K, Pabst A, Luppa M, et al. Validation of MyFORTA: an automated tool to improve medications in older people based on the FORTA list. Drugs Aging. 2024;41(6):555–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Rankin A, et al. Interventions to improve the appropriate use of polypharmacy for older people. Cochrane Database Syst Rev. 2018;9:CD008165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Lee EA, et al. Refining the definition of polypharmacy and its link to disability in older adults: conceptualizing necessary polypharmacy, unnecessary polypharmacy, and polypharmacy of unclear benefit. Perm J. 2020. 10.7812/TPP/18.212. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Khezrian M, et al. An overview of prevalence, determinants and health outcomes of polypharmacy. Ther Adv Drug Saf. 2020;11:2042098620933741. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Varghese D, Patel P. Polypharmacy. In: StatPearls. Treasure Island (FL): StatPearls. Available from: https://www.ncbi.nlm.nih.gov/books/NBK532953/
- 11.Masnoon N, et al. What is polypharmacy? A systematic review of definitions. BMC Geriatr. 2017;17(1):230. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.By the 2023 American Geriatrics Society Beers Criteria® Update Expert Panel. American Geriatrics Society 2023 updated AGS Beers Criteria® for potentially inappropriate medication use in older adults. J Am Geriatr Soc. 2023;71(7):2052–81. 10.1111/jgs.18372. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.O’Mahony D, Cherubini A, Guiteras AR, Denkinger M, Beuscart JB, Onder G, et al. STOPP/START criteria for potentially inappropriate prescribing in older people: version 3. Eur Geriatr Med. 2023;14(4):625–32. 10.1007/s41999-023-00777. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Pazan F, Weiss C, Wehling M, FORTA. The FORTA (Fit fOR The Aged) list 2021: fourth version of a validated clinical aid for improved pharmacotherapy in older adults. Drugs Aging. 2022;39(3):245–7. 10.1007/s40266-022-00922-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Wehling M, Burkhardt H, Kuhn-Thiel A, Pazan F, Throm C, Weiss C, et al. VALFORTA: a randomised trial to validate the FORTA (Fit fOR The Aged) classification. Age Ageing. 2016;45(2):262–7. 10.1093/ageing/afv200. [DOI] [PubMed] [Google Scholar]
- 16.Pazan F, Breunig H, Weiss C, Röhr S, Luppa M, Pentzek M, et al. Higher FORTA (Fit fOR The Aged) scores are associated with poor functional outcomes, dementia, and mortality in older people. Eur J Clin Pharmacol. 2022;78(11):1851–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Pazan F, Burkhardt H, Frohnhofen H, Weiss C, Throm C, Kuhn-Thiel A, et al. Higher Fit-fOR-The-Aged (FORTA) scores comprising medication errors are associated with impaired cognitive and physical function tests in the VALFORTA Trial. Drugs Aging. 2019;36(3):269–77. 10.1007/s40266-018-0626-3. [DOI] [PubMed] [Google Scholar]
- 18.Roehr S, Luck T, Pabst A, et al. Subjective cognitive decline is longitudinally associated with lower health-related quality of life. Int Psychogeriatr. 2017;29(12):1939–50. [DOI] [PubMed] [Google Scholar]
- 19.Hajek A, Brettschneider C, Eisele M, Kaduszkiewicz H, Mamone S, Wiese B, et al. Correlates of hospitalization among the oldest old: results of the AgeCoDe-AgeQualiDe prospective cohort study. Aging Clin Exp Res. 2020;32(7):1295–301. [DOI] [PubMed] [Google Scholar]
- 20.Hajek A, Brettschneider C, Eisele M, et al. Prevalence and determinants of driving habits in the oldest old: results of the multicenter prospective AgeCoDe-AgeQualiDe study. Arch Gerontol Geriatr. 2019;82:245–50. 10.1016/j.archger.2019.03.006. [DOI] [PubMed] [Google Scholar]
- 21.Hajek A, Brettschneider C, Mallon T, et al. Social support and health-related quality of life among the oldest old—longitudinal evidence from the multicenter prospective AgeCoDe-AgeQualiDe study. Qual Life Res. 2022;31(6):1667–76. 10.1007/s11136-021-03070-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Luck T, Riedel-Heller SG, Kaduszkiewicz H, et al. Mild cognitive impairment in general practice: age-specific prevalence and correlate results from the German study on ageing, cognition and dementia in primary care patients (AgeCoDe). Dement Geriatr Cogn Disord. 2007;24(4):307–16. [DOI] [PubMed] [Google Scholar]
- 23.Jessen F, Wiese B, Bickel H, et al. Prediction of dementia in primary care patients. PLoS ONE. 2011;6(2):e16852. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Hajek A, Brettschneider C, Luhmann D, et al. Driving status and health-related quality of life among the oldest old: a population-based examination using data from the AgeCoDe-AgeQualiDe prospective cohort study. Aging Clin Exp Res. 2020;32(7):1295–301. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Roehr S, Luck T, Heser K, et al. Incident subjective cognitive decline does not predict mortality in the elderly-results from the longitudinal German study on ageing, cognition, and dementia (AgeCoDe). PLoS ONE. 2016;11(1):e0147050. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Hajek A, Brettschneider C, Rohr S, et al. Which factors contribute to frailty among the oldest old? Results of the multicentre prospective AgeCoDe and AgeQualiDe study. Gerontology. 2020;66(5):460–6. [DOI] [PubMed] [Google Scholar]
- 27.Hajek A, Brettschneider C, van den Bussche H, et al. Impact of falls on depressive symptoms among the oldest old: results from the AgeQualiDe study. Int J Geriatr Psychiatry. 2018;33(10):1383–8. [DOI] [PubMed] [Google Scholar]
- 28.Luck T, Riedel-Heller SG, Roehr S, et al. Mortality in incident cognitive impairment: results of the prospective AgeCoDe study. J Am Geriatr Soc. 2017;65(4):738–46. [DOI] [PubMed] [Google Scholar]
- 29.Luck T, Pabst A, Roehr S et al. Determinants of incident dementia in different old age groups: results of the prospective AgeCoDe/AgeQualiDe study. Int Psychogeriatr. 2020 May;32(5):645-659. [DOI] [PubMed]
- 30.Buczak-Stec E, Hajek A, van den Bussche H, et al. Frequent attendance in primary care in the oldest old: evidence from the AgeCoDe-AgeQualiDe study. Aging Clin Exp Res. 2020;32(12):2629–38. [DOI] [PubMed] [Google Scholar]
- 31.World Medical A. World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. J Postgrad Med. 2002;48(3):206–8. [PubMed] [Google Scholar]
- 32.Wehling M. Older people, a plethora of drugs, and drug list approaches: useful, efficacious, or a waste of time? J Am Med Dir Assoc. 2016;17(12):1073–5. [DOI] [PubMed] [Google Scholar]
- 33.Schober P, Boer C, Schwarte LA. Correlation coefficients: appropriate use and interpretation. Anesth Analg. 2018;126:1763–8. 10.1213/ANE.0000000000002864. [DOI] [PubMed] [Google Scholar]
- 34.https://www.strobe-statement.org/checklists/. Accessed 24 Jan 2025.
- 35.Rudolf H, Thiem U, Aust K, Krause D, Klaaßen-Mielke R, Greiner W, et al. Reduction of potentially inappropriate medication in the elderly. Dtsch Arztebl Int. 2021;118(51–52):875–82. 10.3238/arztebl.m2021.0372. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Page RL 2nd, Ruscin JM. The risk of adverse drug events and hospital-related morbidity and mortality among older adults with potentially inappropriate medication use. Am J Geriatr Pharmacother. 2006;4(4):297–305. 10.1016/j.amjopharm.2006.12.008. [DOI] [PubMed] [Google Scholar]
- 37.Singh S, Li X, Cocoros NM, Antonelli MT, Avula R, Crawford SL, et al. High-risk medications in persons living with dementia: a randomized clinical trial. JAMA Intern Med. 2024;184(12):1426–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Thiem U, Wilm S, Greiner W, et al. Reduction of potentially inappropriate medication in the elderly: design of a cluster-randomised controlled trial in German primary care practices (RIME). Ther Adv Drug Saf. 2020;12:2042098620918459. 10.1177/2042098620918459. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Ryan C, Teh R, Moyes S, Wilkinson T, Connolly M, Rolleston A, et al. Quality of prescribing predicts hospitalisation in octogenarians: life and living in advanced age: a cohort study in New Zealand (LiLACS NZ). BMC Geriatr. 2019;19(1):357. 10.1186/s12877-019-1305-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.O’Mahony D, Gudmundsson A, Soiza RL, Petrovic M, Cruz-Jentoft AJ, Cherubini A, et al. Prevention of adverse drug reactions in hospitalized older patients with multi-morbidity and polypharmacy: the SENATOR* randomized controlled clinical trial. Age Ageing. 2020;49(4):605–14. 10.1093/ageing/afaa072. [DOI] [PubMed] [Google Scholar]
- 41.Blum MR, Sallevelt BTGM, Spinewine A, O’Mahony D, Moutzouri E, Feller M, et al. Optimizing therapy to prevent avoidable hospital admissions in multimorbid older adults (OPERAM): cluster randomised controlled trial. BMJ. 2021;374:n1585. 10.1136/bmj.n1585.39. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Michalek C, Wehling M, Schlitzer J, Frohnhofen H. Effects of “Fit fOR The Aged” (FORTA) on pharmacotherapy and clinical endpoints—a pilot randomized controlled study. Eur J Clin Pharmacol. 2014;70(10):1261–7. [DOI] [PubMed] [Google Scholar]
- 43.Azab M, Novella A, Pasina L. FORTA score and negative outcomes in older adults: insights from Italian internal medicine wards. Med Princ Pract. 2025;34(1):57–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Chen LJ, Nguyen TNM, Laetsch DC, Chang-Claude J, Hoffmeister M, Brenner H, et al. Association of comedication quality with chemotherapy-related adverse drug reactions and survival in older colorectal cancer patients. J Gerontol A Biol Sci Med Sci. 2022;77(5):1009–19. [DOI] [PubMed] [Google Scholar]
- 45.Degen M, Haug U, Scholle OHF, Awuku-Larbi Y, Carlsson S, Ebert N, Harth V, Hauri AM, Heier M, Herrmann A, Leitzmann M, Linseisen J, Meinke-Franze C, Obi N, Pendt P, Reinhold T, Thorand B, Brenner H, Schöttker B. Factors associated with potentially inappropriate medication use, medication underuse and overuse in older adults in the German National Cohort. J Intern Med. 2026. 10.1111/joim.70097. Epub ahead of print. PMID: 41964402. [DOI] [PMC free article] [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Upon submission of a written request and evaluation by the study group, the data used in this study will be provided to other researchers.