Systematic review and critical methodological appraisal of community-based falls prevention economic models

Table 4 Time-variant falls risk factors and determinants of background health state utility and care cost progressions in non-binary models with horizons longer than 5 years

Study label^a	Time-variant falls risk factors			Background^b health/cost determinants
Study label^a	Age	Falls incidence	Other	Health state utility	Comorbidity care cost
Cohort-level Markov models^c
Boyd et al. (2020)	Tunnel state?^d	MA fall		Age^e	Age^e
Church et al. (2011, 2012)	Tunnel state?	Any fall		Age^f
Deverall et al. (2018)	Tunnel state?	MA fall		Age^e	Age^e
Eldridge et al. (2005)	Tunnel state?	Fracture	FoF	Unclear^f	Post-fracture^g
Farag et al. (2015)	Tunnel state?	Any fall		Age^f
Honkanen et al. (2006)	Tunnel state?	Hip fracture	FS	Age^e; FS; LTC; post-hip fracture	FS; LTC
Johansson et al. (2008)	Tunnel state?	Hip fracture		Age^e,f	Age
Moriarty et al. (2019)	Tunnel state?	MA fall, Hip fracture		Age^f
OMAS (2008)	Tunnel state?	MA fall		CEA
Pega et al. (2016)	Tunnel state?	MA fall		Age^e	Age^e
RCN (2005)	Tunnel state?			Age
Wilson et al. (2017)	Tunnel state?	MA fall		Age^e	Age^e
Individual-level Markov models
Hiligsmann et al. (2014)	×			Post-hip and vertebral fractures	Post-hip fracture^g
Mori et al. (2017)	×	Fracture	Osteoporosis	Age; post-hip and vertebral fractures	Post-hip fracture^g
Nshimyumukiza et al. (2013)	×	Fracture	BMD	Post-hip and vertebral fractures
Zarca et al. (2014)	×	Hip fracture	Vitamin D	Age; post-hip fracture

BMD bone mass density, CEA cost-effectiveness analysis, FoF fear of falling, FS functional status, LTC long-term care, MA fall fall requiring medical attention, OMAS Ontario Medical Advisory Secretariat, RCN Royal College of Nursing
^aSee Table 1 for study references
^bNot directed related to but indirectly influenced by falls/fractures: e.g., fatal fall influences lifetime comorbidity care costs
^cAlhambra-Borras et al. [146] was excluded due to unclear description of the dynamic model states following intervention
^dAge-based risk progression would require tunnel states but this was not mentioned or graphed, hence the question mark
^eStratified by further time-invariant factors including sex and ethnicity
^fUnclear whether events such as fracture and LTC admission incurred a one-off or permanent health utility loss
^gIncorporated ongoing care costs for serious fractures, which are not technically comorbidity care costs since they are directly associated with fall/fracture in model; but they can be interpreted as such given their permanent nature

ISSN: 1478-7547