The Joint Action (JA) on Frailty Prevention, called JAHEE, aims to develop a holistic
and comprehensive strategic framework for the prevention and management of frailty at
European level. Although frailty is increasing worldwide, its prevalence at population level in
Europe is unclear and little is known about frailty incidence, trajectories or transitions
between different stages of frailty, or the feasibility, consequences and potential benefits
associated with screening, surveillance and monitoring for frailty at population level.

Objectives:

To investigate (1) the current epidemiological data (prevalence, incidence and
trajectories/transitions) concerning frailty and (2) the evidence supporting the
development, implementation and evaluation of population-based approaches to screening,
surveillance and monitoring of frailty, including those delivered in the primary care setting,
in 22 European Union (EU) JA JAHEE Member States (MSs).

Methods:

We systematically reviewed the published literature using the PubMed, Embase, CINAHL,
MEDLINE, Opengrey and the Cochrane Library databases from January 2002 through to April
2017. Good practices and grey literature from ongoing or unpublished European-supported
frailty projects were also sought from JA JAHEE MS partners. Selected studies were
those that explicitly addressed the epidemiology of frailty and population-based approaches
in adults (≥18) in any settings in JA JAHEE MSs to investigate the prevalence and
incidence of frailty, trajectories and transitions between different stages of frailty and the
current practice of screening, surveillance and monitoring for frailty.

Results:

Five systematic reviews were conducted. A total of 5,030 abstracts were screened
independently by two assessors. After eligibility assessment, 69 full papers were analysed to
extract data relating to the five topics addressed by the systematic reviews:

(1) Prevalence and incidence; n=63 papers,
(2) Screening; n=3,
(3) Surveillance; n=0,
(4) Monitoring; n=0 and
(5) Trajectories and transitions; n=3.

Prevalence of frailty varied between settings, and by age range and frailty assessment
instrument used. Most studies reporting prevalence data were from community-based samples (n=53) with only five studies available from primary care. Studies from Spain (n=11)
and the Netherlands (n=11) were best represented in the literature. The main frailty
classification used was the Frailty phenotype based on the Cardiovascular Health Study (56%
of all studies). The highest prevalence rates were from studies set in hospital wards (around
50%) and Long-Term Care (≥60%).

Lower rates, around 30%, were found in studies of adults>18 years in primary care and outpatient settings. The prevalence reported in community samples ranged from 2-60% with most reporting rates <30%. No prevalence data were available for five JA JAHEE countries (Bulgaria, Croatia, Cyprus, Lithuania or Malta). Prevalence largely increased with age irrespective of settings. Three studies reported on frailty incidence but used incidence proportions rather than rates; sample characteristics and follow-up length were heterogeneous, limiting comparability.

Two studies meeting inclusion criteria reported approaches to screen for frailty; both used a two-step screening and assessment method. Both were conducted in the community, one in primary care and the other in patients’ homes. No study from a JA JAHEE MS described a systematic process for surveillance or monitoring of frailty. Studies presenting data demonstrating frailty transitions or trajectories in JA JAHEE MSs (n=3) were also heterogeneous in design (sample, definition of frailty, duration of follow-up and outcomes, etc.) making comparison challenging.

Recommendations for JA JAHEE:

Prevalence and incidence of frailty varied by study setting, design, population features and
frailty classification used. Few papers reported data from primary care. The lack of data for
population-level approaches to screening, surveillance and monitoring of frailty, or on frailty
transitions, limits conclusions.

The extent of the heterogeneity between studies in terms of
setting, age range, gender balance, follow-up duration and frailty classification needs to be
addressed to provide more reliable and comparable estimates of frailty incidence and
prevalence and to develop and harmonise data sets across JA JAHEE MSs.