The Fit fOR The Aged (FORTA) project and its clinical implications

KEYWORDS:

• Fit fOR The Aged (FORTA)
• older people
• inappropriate prescribing
• potentially inappropriate medication (PIM)
• deprescribing
• drug-oriented listing approach (DOLA)
• patient-in-focus listing approach (PILA)
• medication optimization
• drug therapy

1. Introduction

The demographic change of aging societies around the globe [1] is associated with an increasing prevalence of chronic illnesses among older people (65 years and older) rendering the use of multiple drug treatments in this population often inevitable. Unfortunately, evidence on safety and efficacy is still lacking or sparse for the use of many drugs to treat older patients [2]. Yet, treatment has to be applied even in the absence of compelling evidence [from randomized controlled trials (RCT), for example] and, thus strict guidance by evidence-based guidelines. Paying tribute to this problem, several approaches have been developed to increase the appropriateness of drug treatment in older people [3]. A widely applied model is the development of drug lists labeling the appropriateness of drugs for this population; in most cases drug assessment is based on the combination of existing evidence and experts’ opinions in a consensual procedure (Delphi process, for example).

The resulting lists are impressively diverse: they may label drugs both positively (should be used) or negatively (to be avoided) (FORTA as an example), or provide both recommendations on drugs and actions (e.g. START criteria), or compile generic questions, but provide no answers regarding the appropriateness of drug treatment (e.g. MAI) [3]. Besides, these approaches may differ regarding recommendations pertaining various aspects of drug therapy such as drug-drug interactions, dosing and kidney function.

2. FORTA and the realm of listing approaches

For better orientation, we recently proposed to classify the 73 listing approaches found in a recent systematic review either as drug-oriented listing approaches (DOLA) such as the Beers criteria® which is a list of potentially inappropriate medications for older people [4] or patient-in-focus listing approaches (PILA) such as FORTA [3]. Only 9 of these approaches were PILA. As expected, only few randomized controlled trials with relevant clinical outcomes such as adverse drug reactions (ADRs), falls, hospitalization, pain or mortality as an endpoint (N = 13) were found; regarding the outcomes of endpoints PILA such as the START/STOPP criteria which mainly consist of actions and partly of medications to be given or avoided in older patients [5] or FORTA were superior to DOLA [3]. Key differences between PILA and DOLA are depicted in Figure 1.

Figure 1. Comparison between patient-in-focus listing approaches (PILA, left) and drug-oriented listing approaches (DOLA, right).

3. The FORTA classification and the development of the FORTA lists

The FORTA classification was introduced in 2008 as a clinical aid for physicians to improve drug treatment in geriatric patients. FORTA classifies medications with regard to their safety, efficacy and overall age-appropriateness into four categories ranging from A-B-C to D [6,7]. FORTA-A drugs can be generally regarded as indispensable based on their proven benefits in terms of efficacy and safety, FORTA-B drugs are beneficial but also have some limitations with regard to safety and efficacy, FORTA-C drugs have a questionable safety/efficacy profile, require intense monitoring and should be avoided in the presence of too many drugs; FORTA-D drugs should be generally avoided [7]. In 2012, these labels were used to develop an author-based list of frequently used medications for common diseases including cardiovascular, metabolic, oncological and psychiatric disorders [6]. This list contained 190 medications aligned to 20 main indication groups; it was used as a proposal for the development of the first consensus validated version of the FORTA list (FORTA 2012) in Germany and Austria [6] finally including 225 drug/drug groups belonging to 24 indications. Encouraged by the results of the clinical trials (see section 4) seven country-region-specific FORTA lists and a European FORTA list (containing 264 items in 26 main indication groups) were developed [7]. Recently, the third version of FORTA list for German speaking countries containing 296 entries in 30 indications as well as the first U.S.-FORTA list containing 273 items aligned to 27 major indication groups were published [8]. The Japanese FORTA list awaits publication, work on an Australian list is ongoing.

Two specialty FORTA assessments have been performed: in the OAC-FORTA list [9] oral anticoagulants have been labeled (2^nd version 2019, submitted) from C (less studied vitamin-K-antagonists such as phenprocoumon) through B (warfarin, edoxaban, rivaroxaban, dabigatran) to A (apixaban). In LUTS-FORTA, oral drugs to treat urinary tract syndromes, mainly incontinence, were assessed [10]. In both cases, participating raters were more specialized than those generalists on the rater panel for the long FORTA list.

4. Clinical validation of FORTA

In the VALFORTA trial [11], the use of the first consensus validated FORTA list based on the FORTA method [12] led to a significant improvement of medication quality. This was measured by the FORTA score representing the individual sum of over- and under-treatments as identified by negatively (overtreatment) or positively (undertreatment) labeled drugs to be removed or added. Briefly, the reduction of the FORTA score and thus improvement of medication quality in the FORTA intervention group was 2.7 fold greater than in the control group (p < 0.0001) [11]. The VALFORTA trial was the first larger RCT on a listing approach to yield in the improvement of relevant clinical endpoints such as activities of daily living or the occurrence of adverse drug reactions; the number needed to treat for latter effect was only 5, meaning that one side effect was gone after treating only 5 patients according to FORTA. Moreover, secondary analyses of the VALFORTA trial revealed that the application of FORTA also significantly improved the medication quality at the diagnosis-related, individual drug/drug class level and that higher FORTA scores are associated with worse functional outcomes [13,14]. This study did not only demonstrate the clinical effectiveness of FORTA but also its teachability, implementability and, thus, utility in a clinical setting. The intervention consisted in information on FORTA, basic training of doctors and weekly FORTA drug review boards on the interventional wards. This type of intervention would be easily implementable to start the use of FORTA in a clinical setting. An implementation in an ambulatory setting has not yet been studied though a recent survey in the state of Baden-Württemberg, Germany, showed that the current dissemination strategy has introduced FORTA into about half of the responding practitioners with about 1/3 using it regularly (unpublished results).

5. Computerization of FORTA

To facilitate the use of the FORTA list in daily clinical practice, the FORTA App was launched in 2017 and recently updated. The actual version includes the latest FORTA classifications from the FORTA list 2018. Besides, a novel information technology algorithm (The FORTA-EPI Algorithm) was developed to assess the quality of medication according to FORTA based on bulk patient data from insurance companies [15]. This algorithm allows for automated analysis of large datasets regarding medication quality and may be used in epidemiological research.

6. Future perspectives

The FORTA algorithm is being developed into an individualized tool that factors patient response biomarkers (e.g. blood pressure or pain level) into the assessment.

Further prospective clinical trials on the long-term impact of FORTA on relevant clinical endpoints in different settings (clinics, outpatients) are planned, but await funding.

Extensive teaching, best practice seminars and further educational activities are measurably increasing knowledge on and the use of FORTA and are necessary for the wider implementation of FORTA internationally.

7. Expert opinion

Polypharmacy and inappropriate drug therapy are highly prevalent and pose a threat to older people. There is an abundance of listing approaches to tackle the problem of inappropriate drug treatment. A major discriminator for these lists is the degree of consideration of patients` characteristics and demands. In addition, the majority of the existing listing approaches have not been validated in randomized clinical trials (RCT). Based on the few existing RCT, listing approaches that omit patient’s characteristics and demands and just focus on deleting inappropriate medications are inferior to those taking patients` characteristics and demands into account (e.g. FORTA). FORTA represents the first positive/negative labeling approach at the individual drug level resulting in a drug list; this list has been shown to improve the quality of medication as well as relevant clinical outcomes in an RCT. Therefore, its use and that of other prospectively validated PILA should be preferred over that of DOLA.

Declaration 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 Sanofi-Aventis, Bayer, Berlin-Chemie, Boehringer-Ingelheim, Aspen, Novartis, Takeda, Roche, Pfizer, Bristol-Myers, Daichii-Sankyo, Lilly, Otsuka, Novo-Nordisk, Shire, and LEO Pharma. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer Disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This paper was not funded.